Search

The use of Artificial Intelligence (AI) to preserve and promote cultural heritage has experienced significant growth in recent years. Among the various areas of cultural heritage, numismatics have emerged as a particularly promising field where we can develop AI solutions. Numismatics refers to the study of coins, tokens, paper money, and medals, which play a critical role in understanding human history and culture. However, there are still limited resources available to help researchers and collectors in the identification of coins. This is due to the vast number of coins in circulation, which presents a significant challenge in developing smart tools for classification tasks. This paper aims to provide a contribution to this setting. In particular, we start by creating a new dataset called EURO-Coin, which consists of images showing the side of coins with reliefs and is designed to facilitate the training and testing of AI models for euro coin classification. Then, we propose two approaches that leverage Convolutional Neural Networks and self-attention layers to classify the country and value of the coins. In our experiments, we obtain an accuracy of 86.9% for country classification and an accuracy of 96.4% for value classification. Finally, we conduct an ablation study to evaluate the impact of the preprocessing activities and attention layers in our approaches. [ABSTRACT FROM AUTHOR]

This paper presents a convolutional neural network (CNN) to classify between the conventionally and organically cultivated Memecik varieties of green olives. The image forming method called the rising paper chromatography is utilized in preparing the images of Memecik varieties of green olives for CNN. In the rising chromatography method, 20, 30, and 40% sample concentrations were determined as the suitable concentrations for both organic and conventional olives. The concentrations of AgNO3 and FeSO4 were determined as 0.25, 0.5, 0.75 and 1% for both conventional and organic samples. The visual differences used for differentiation of different types of Memecik green olives are usually determined according to the regional color differences, the vivid color occurrence, the width and the frequency of bowl occurrence, the thin line, and the picks at drop zone by the expert assessors. The testing results in this study verified the effectiveness of the CNN methodology in differentiating between the organically and conventionally cultivated Memecik green olives. The newly designed neural network achieved 100% accuracy. Furthermore, this high accuracy achieved by CNN might suggest that it can be effectively used in place of the expert assessors. [ABSTRACT FROM AUTHOR]

Traditional sentiment analysis focuses on text-level sentiment mining, transforming sentiment mining into classification or regression problems, resulting in a sentiment analysis low accuracy rate. Sentiment analysis refers to the use of natural language processing, text analysis, and computational linguistics to systematically identify, extract, quantify, and study sentimental states. Therefore, more scholars have begun to focus on speech recognition and facial expression recognition research, and extracting and analysing people's sentiment tendencies can improve sentiment recognition accuracy. Traditional single-modal sentiment analysis can no longer meet people's needs. Therefore, this paper proposes a multimodal sentiment analysis method based on the multimodal sentiment analysis method that can obtain more sentimental information sources and help people make better decisions. The experimental results in this paper show that the highest recognition rates of CNN-SVM, RNN-SVM, and CRNN-SVM were 76.8%, 71.2%, and 93.5%, respectively. It can be seen that CRNN-SVM has the highest sentiment tendency recognition rate in deep learning, so it is suitable to apply CRNN-SVM to sentiment tendency analysis system design in this paper. The average accuracy rate of the system designed in this paper was 91%, and the stability was also very strong, which shows that the system designed in this paper is meaningful. The main contribution of this paper is based on the limitations of single-mode emotion analysis. It proposes a multimode emotion analysis method and introduces a convolutional neural network to help people obtain more emotional information sources to meet their needs. [ABSTRACT FROM AUTHOR]

Brain–computer interfacing (BCI) research, fueled by deep learning, integrates insights from diverse domains. A notable focus is on steady-state visual evoked potential (SSVEP) in BCI applications, requiring in-depth assessment through deep learning. EEG research frequently employs SSVEPs, which are regarded as normal brain responses to visual stimuli, particularly in investigations of visual perception and attention. This paper tries to give an in-depth analysis of the implications of deep learning for SSVEP-adapted BCI. A systematic search across four stable databases (Web of Science, PubMed, ScienceDirect, and IEEE) was developed to assemble a vast reservoir of relevant theoretical and scientific knowledge. A comprehensive search yielded 177 papers that appeared between 2010 and 2023. Thence a strict screening method from predetermined inclusion criteria finally generated 39 records. These selected works were the basis of the study, presenting alternate views, obstacles, limitations and interesting ideas. By providing a systematic presentation of the material, it has made a key scholarly contribution. It focuses on the technical aspects of SSVEP-based BCI, EEG technologies and complex applications of deep learning technology in these areas. The study delivers more penetrating reporting on the latest deep learning pattern recognition techniques than its predecessors, together with progress in data acquisition and recording means suitable for SSVEP-based BCI devices. Especially in the realms of deep learning technology orchestration, pattern recognition techniques, and EEG data collection, it has effectively closed four important research gaps. To increase the accessibility of this critical material, the results of the study take the form of easy-to-read tables just generated. Applying deep learning techniques in SSVEP-based BCI applications, as the research shows, also has its downsides. The study concludes that a radical framework will be presented which, includes intelligent decision-making tools for evaluation and benchmarking. Rather than just finding a comparable or similar analogy, this framework is intended to help guide future research and pragmatic applications, and to determine which SSVEP-based BCI applications have succeeded at responsibility for what they set out with. [ABSTRACT FROM AUTHOR]

Marine object detection has become increasingly important in intelligent underwater robot. Because of color cast and blur in underwater images, features directly extracted from backbone networks usually lack interesting and discriminative characters, that affects performance on marine object detection. To this end, this paper proposes a novel refined marine object detector with attention-based spatial pyramid pooling networks and bidirectional feature fusion strategy to relieve the weakening of features and address marine object detection issues. Firstly, an attention-based spatial pyramid pooling network named as SA-SPPN is proposed to enrich interesting information and extend receptive field on original features extracted from backbone network. Based on enhanced multiple level features, the bidirectional feature fusion strategy is designed to fuse different level features and generate robust feature maps for detection. Specifically, the top-down connection could transfer semantic information from high-level features to enhance low-level features. The bottom-up pathway could extend resolution of high-level features. Furthermore, the cross-layer connections are integrated into both top-down passway and bottom-up passway to carry out multiple branch fusion. On bounding boxes regression phase, the distance-IoU loss is adopted to improve regression speed and accuracy. Finally, this paper conducts series experiments on underwater image datasets and URPC datasets to detect marine objects. The experimental results reveal that our approach could achieve impressive performance and reach 79.64% mAP on underwater image datasets, 79.31% mAP on URPC2019 datasets and 79.93% mAP on URPC2020 datasets, respectively. For standard object detection, the proposed algorithm also could realize notable performance and get 81.9% mAP on PASCAL VOC datasets. [ABSTRACT FROM AUTHOR]

Chinese calligraphy is an important part of Chinese national culture and art and part of the essence of Chinese national culture. Stone calligraphy is one of the important elements of Chinese calligraphy art. Stone carved calligraphy characters have high cultural and artistic value. Therefore, accurately recognizing stone carved calligraphy characters are of great importance. Stone carved calligraphy can identify hard-to-preserve stone calligraphy paper materials in electronic data that can be preserved for a long time, thereby offering important reference materials for the study of the historical development of Chinese calligraphy art. Moreover, with the development of science and technology and the investment of China in cultural and artistic undertakings, computer-aided calligraphy character recognition technology is also constantly improving, and its application in calligraphy recognition is becoming increasingly extensive. This article aims to study a method of stone inscription calligraphy recognition based on convolutional neural networks. In this paper, we use an image recognition and optimization method consisting of a convolutional neural network to carry out an experiment with stone inscription calligraphy characters. It was concluded that the recognition accuracy of stone calligraphy characters by the convolutional neural network reached 99.2%, indicating that this stone calligraphy character recognition method based on a convolutional neural network has a good ability to recognize stone calligraphy characters. [ABSTRACT FROM AUTHOR]

High-resolution remote sensing images have become mainstream remote sensing data, but there is an obvious "salt and pepper phenomenon" in the existing semantic segmentation methods of high-resolution remote sensing images. The purpose of this paper is to propose an improved deep convolutional neural network based on HRNet and PSPNet to segment and realize deep scene analysis and improve the pixel-level semantic segmentation representation of high-resolution remote sensing images. Based on hierarchical multiscale segmentation technology research, the main method is multiband segmentation; the vegetation, buildings, roads, waters and bare land rule sets in the experimental area are established, the classification is extracted, and the category is labeled at each pixel in the image. Using the image classification network structure, different levels of feature vectors can be used to meet the judgment requirements. The HRNet and PSPNet algorithms are used to analyze the scene and obtain the category labels of all pixels in an image. Experiments have shown that artificial intelligence uses the pyramid pooling module in the classification and recognition of CCF satellite images. In the context of integrating different regions, PSPNet affects the region segmentation accuracy. FCN, DeepLab and PSPNet are now the best methods and achieve 98% accuracy. However, the PSPNet object recognition algorithm has better advantages in specific areas. Experiments show that this method has high segmentation accuracy and good generalization ability and can be used in practical engineering. [ABSTRACT FROM AUTHOR]

In this paper, we have proposed a hybrid bio-inspired algorithm which takes the merits of whale optimization algorithm (WOA) and adaptive particle swarm optimization (APSO). The proposed algorithm is referred as the hybrid WOA_APSO algorithm. We utilize a convolutional neural network (CNN) for classification purposes. Extensive experiments are performed to evaluate the performance of the proposed model. Here, pre-processing and segmentation are performed on 120 lung CT images for obtaining the segmented tumored and non-tumored region nodule. The statistical, texture, geometrical and structural features are extracted from the processed image using different techniques. The optimized feature selection plays a crucial role in determining the accuracy of the classification algorithm. The novel variant of whale optimization algorithm and adaptive particle swarm optimization, hybrid bio-inspired WOA_APSO, is proposed for selecting optimized features. The feature selection grouping is applied by embedding linear discriminant analysis which helps in determining the reduced dimensions of subsets. Twofold performance comparisons are done. First, we compare the performance against the different classification techniques such as support vector machine, artificial neural network (ANN) and CNN. Second, the computational cost of the hybrid WOA_APSO is compared with the standard WOA and APSO algorithms. The experimental result reveals that the proposed algorithm is capable of automatic lung tumor detection and it outperforms the other state-of-the-art methods on standard quality measures such as accuracy (97.18%), sensitivity (97%) and specificity (98.66%). The results reported in this paper are encouraging; hence, these results will motivate other researchers to explore more in this direction. [ABSTRACT FROM AUTHOR]

Image classification can effectively manage and organize images, laying a good foundation for the work in multiple fields of image processing. With the rise of Internet technology and social networks, the number of digital images has increased dramatically and there are more and more applications. People also use intuitive pictures instead of words when expressing emotions and information. A large number of digital images need to be managed, analyzed, and retrieved. This urgently requires more efficient and accurate image classification technology. Deep learning is a learning method that extends traditional neural networks, simulating the process of gradual abstraction of human brain cognition. The number of hidden layers is deepened, and features can be learned automatically. This paper has completed the traditional image noise detection and segmentation experiment based on convolutional neural network. We introduced the various parts of the convolutional neural network, including the convolutional layer, activation function, fully connected layer, etc. Noise detection is completed based on Fast RCNN on the MSTAR data set containing the background. The model fully combines the advantages of the isotropic model, the Perona-Malik model, and the second-order directional derivative. In order to better maintain the edge structure information of the image, the structure information is extracted based on the original noise image, and the improved ID model and the improved PM model are directionally diffused along the edge tangent direction of the original noise image. Considering the local structure information of the image, we use the slice similarity modulus value as the edge detector of the improved PM model, and use the slice similarity modulus value to construct a new weighting function to adaptively balance the relative weights of the improved ID model and the improved PM model. Simulation and measured data verify the effectiveness of this network in removing image coherent speckle noise. We compare and analyze it with existing denoising methods. The use of visual evaluation and objective evaluation indicators to evaluate the denoising effect and calculation efficiency shows the advantages of the network in this paper in terms of denoising effect, calculation time and space complexity. [ABSTRACT FROM AUTHOR]

In an iron ore pelletization plant, pellets are produced inside a rotating disc pelletizer. Online pellet size distribution is an important performance indicator of the pelletization process. Image processing-based system is an effective solution for online size analysis of iron ore pellets. This paper proposes a machine learning algorithm for estimating the size class of the pellets during their production by imaging from an area inside the disc pelletizer. Instead of computing the size of each individual pellets in the acquired image, this method proposes a qualitative approach to get the overall size estimate of the pellets in production. The key idea of this paper is to find out whether the disc is producing VERY SMALL, SMALL, MEDIUM, or BIG-sized pellets. A weighted average ensemble of different convolutional neural networks such as VGG16, Mobilenet, and Resnet50 is used to achieve this objective. Furthermore, batch normalization is applied to improve the estimation performance of the proposed model. A novel data augmentation method is applied to the in situ captured images to create the data set used to train and evaluate the proposed ensemble of CNN models. Results of experiments indicate that it is possible to detect the operating state of the pelletization disc by acquiring images from the inside area of the disc with sufficient accuracy. [ABSTRACT FROM AUTHOR]

This paper analyzes the inherent relationship between convolutional neural networks and sparse representation and proposes an improved convolutional neural network model for image synthesis in response to problems with current methods. In the testing phase, the calculation of the sparse coefficients involves the solution of complex optimization problems, which greatly reduce the operating efficiency, inspired by the successful application of convolutional neural networks in the field of image reconstruction. Compared with the traditional image portrait synthesis method, this model not only has an end-to-end closed form but also does not need to solve complex optimization problems in the synthesis stage. The synthesis experiment on an image dataset shows that this method not only improves the synthesis effect but also improves the efficiency of the traditional method by one to two orders of magnitude, demonstrating its potential application value. Blocking processing is a common method for sparse domain image modeling. It improves the computational efficiency but also decreases the global structure of the image, which is difficult to compensate for through the aggregation and overlap of image blocks. In response to this problem, this paper proposes a low-rank image inpainting method based on a Gaussian mixture model. This method embeds the local statistical characteristics of image blocks into the kernel norm model and not only uses the Gaussian mixture model to maintain the local details of the image but also describes the global low-rank structure of the image through the kernel norm, thus restoring a class of image data with a potential low-rank structure and theoretically revealing the structured sparse nature of the Gaussian mixture model. This paper optimizes the strategy based on random hidden neuron nodes and proposes a dropout anti-overfitting strategy based on sparsity. The experiments show that this strategy can effectively improve the convergence speed while ensuring good performance and can effectively prevent overfitting. [ABSTRACT FROM AUTHOR]

The path planning system is an important part of unmanned vehicles, and the development of path planning technology will surely promote the rapid development of unmanned vehicle technology. In order to prevent the node from continuously monitoring its state, a self-triggering control strategy is proposed. Before the trigger moment, the node does not need to monitor its state. Moreover, considering the unpredictable problem of the node state, a control strategy triggered by the observed event is proposed, that is, only the output state information is used to determine the trigger time. In addition, this paper analyzes and models the two major factors that affect the local planning results, the environment and the vehicle, and uses the path smoothing and optimization method based on B-spline curve and the path optimization method based on the steering controller. Finally, this paper designs experiments to analyze the vehicle local path planning method and time consistency of the unmanned driving system. From the experimental results, it can be seen that the unmanned driving system constructed in this paper has a certain effect. [ABSTRACT FROM AUTHOR]

Microscopic image analysis technology helps solve the inadvertences of artificial traditional methods in disease, wastewater treatment, and environmental change monitoring analysis. Convolutional neural network (CNN) play an important role in microscopic image analysis. Image segmentation, in which U-Net is increasingly applied in microscopic image segmentation, is a crucial step in detection, tracking, monitoring, feature extraction, modelling, and analysis. This paper comprehensively reviews the development history of U-Net, analyses several research results of various segmentation methods since the emergence of U-Net, and conducts a comprehensive review of related papers. This paper summarised the improved methods of U-Net and then listed the existing significance of image segmentation techniques and their improvements introduced over the years. Finally, focusing on the different improvement strategies of U-Net in different papers, the related work of each application target is reviewed according to detailed technical categories to facilitate future research. Researchers can see the dynamics of the transmission of technological development and keep up with future trends in this interdisciplinary field. [ABSTRACT FROM AUTHOR]

With the advancement in modern imaging techniques like CT scan, MRI, PET scan etc., a vast amount of data is generated every day in the field of healthcare. Big data contains hidden information, which necessitates the development of intelligent systems to analyze it and extract relevant information, allowing for accurate and cost-effective decisions in the medical field. By utilizing the untapped potential of the big data available in the medical field, very precise models can be developed for the medical diagnosis of retinal diseases. Optical coherence tomography (OCT) is a non-invasive imaging test that captures different, distinctive layers of the retina and optic nerve in a living eye to map and measure their thickness, that helps diagnose various retinal disorders. With the advancement of the application of deep learning-based techniques in the field of medical sciences, the use of convolutional neural network (CNN) based approaches for disease detection is gaining popularity. While the manual examination of 3D OCT images for the diagnosis of retinal disorders requires extensive time and expert intervention, the use of CNNs provides an effective automated option that provides results with higher accuracy while also reducing the time involved in the overall process. In this paper, we have implemented the aforementioned idea by proposing OCT-CNN, a CNN architecture, that automatically classifies retinal OCT images and identifies potential disorders in a living eye. Several techniques have been employed to enhance the performance of the proposed approach, including digital enhancement of the images, dropout regularization, adaptive learning rates, and early stopping of training to attain optimal performance. The performance of the proposed OCT-CNN is evaluated on the UCSD dataset against several popular deep CNN architectures and existing state-of-the-art approaches to automatic retinal OCT classification. The proposed OCT-CNN attains the best performance on all evaluated metrics, pushing the classification accuracies to 99.28% on CNV, 99.9% on DME, 99.38% on DRUSEN, and 100% on NORMAL images, indicating its superiority over existing state-of-the-art techniques. [ABSTRACT FROM AUTHOR]

Dry eye disease (DED) is a chronic eye disease and a common complication among the world's population. Evaporation of moisture from tear film or a decrease in tear production leads to an unstable tear film which causes DED. The tear film breakup time (TBUT) test is a common clinical test used to diagnose DED. In this test, DED is diagnosed by measuring the time at which the first breakup pattern appears on the tear film. TBUT test is subjective, labour-intensive and time-consuming. These weaknesses make a computer-aided diagnosis of DED highly desirable. The existing computer-aided DED detection techniques use expensive instruments for image acquisition which may not be available in all eye clinics. Moreover, among these techniques, TBUT-based DED detection techniques are limited to finding only tear film breakup area/time and do not identify the severity of DED, which can essentially be helpful to ophthalmologists in prescribing the right treatment. Additionally, a few challenges in developing a DED detection approach are less illuminated video, constant blinking of eyes in the videos, blurred video, and lack of public datasets. This paper presents a novel TBUT-based DED detection approach that detects the presence/absence of DED from TBUT video. In addition, the proposed approach accurately identifies the severity level of DED and further categorizes it as normal, moderate or severe based on the TBUT. The proposed approach exhibits high performance in classifying TBUT frames, detecting DED, and severity grading of TBUT video with an accuracy of 83%. Also, the correlation computed between the proposed approach and the Ophthalmologist's opinion is 90%, which reflects the noteworthy contribution of our proposed approach. [ABSTRACT FROM AUTHOR]

With the leaps and bounds of computer performance and the advent of the era of big data, deep learning has drawn more and more attention from all walks of life. It can combine low-level features to form more abstract high-level features and describe the data more essentially. Therefore, it is widely used in various fields such as computer vision. Image segmentation is one of the most basic research topics in the field of computer vision. The main purpose is to extract regions of interest from images for later image processing research. In 3D reconstruction based on sequence images, the segmentation accuracy and speed of sequence images determine the quality and efficiency of target reconstruction. Therefore, when facing large-scale sequence images, the biggest problem is how to improve the segmentation speed while ensuring accuracy. Based on the above background, the research content of this article is an indoor scene segmentation algorithm based on full convolutional neural network. According to the characteristics of indoor application scenes, this paper proposes a fast convolutional neural network image segmentation method to segment the indoor scene image and construct the fast fully convolutional networks (FFCN) for indoor scene image segmentation uses inter-layer fusion to reduce the amount of network calculation parameters and avoid the loss of picture feature information by continuous convolution. In order to verify the effectiveness of the network, in this paper, a basic living object data set (XAUT data set) in an indoor environment is created. The XAUT data set is used to train the FFCN network under the Caffe framework to obtain an indoor scene segmentation model. In order to compare the effectiveness of the model, the structure of the worn FCN8s, FCN16s, and FCN32s models was fine-tuned, and the corresponding algorithm model for indoor scene segmentation was obtained by training with the XAUT data set. The experimental results show that the pixel recognition accuracy of all types of networks has reached 86%, and the mean IU ratio has reached more than 63%. The mean IU of the FCN8s network is the highest at 70.38%, but its segmentation speed is only 1/5 of FFCN. On the premise that other types of indicators are not much different, the average segmentation speed on FFCN fast segmentation convolutional neural network reaches 40 fps. It can be seen that the scale fusion technology can well avoid the loss of image feature information in the network convolution and reddening process. Compared with other FCN networks, it has a faster speed and is conducive to real-time image preprocessing. [ABSTRACT FROM AUTHOR]

Scene recognition has been the foundation of research in computer vision fields. Because scene images typically are composed of specific regions distributed in some layout, so modeling layouts of various scenes is a key clue for scene recognition. Existing methods usually require an additional stream to detect regions for subsequent modeling, which accumulate errors and may miss important information. Meanwhile, they use manual features to model relations between regions, which weakens the representation ability of layouts. In this paper, we propose a single-stream adaptive scene layout modeling approach based on a layout modeling module (LMM), which constructs layouts without additional detection streams and adaptively captures the relations to take advantage of graph attention network. LMM is directly concatenated to a convolutional neural network, where each pixel of the activation maps of the last convolutional layer is defined as a region that is the initial input node of the LMM. LMM first models the layout of each region, and then uses all regions with layout information to model the entire scene. Layout relations are encoded as edges, which are automatically analyzed according to region co-occurrence and relative position. Our work can be understood as optimizing features of the activation maps from a scene layout modeling perspective for scene recognition. Experimental results on MIT67, SUN397, and Places365 show that our single-stream model achieves competitive performance. [ABSTRACT FROM AUTHOR]

The purpose of smart city is to enhance the optimal utilization of scarce resources and improve the resident's quality of live. The smart cities employed Internet of Things (IoT) to create a sustainable urban life. The IoT devices such as sensors, actuators, and smartphones in the smart cities generate data. The data generated from the smart cities are subjected to analytics to gain insight and discover new knowledge for improving the efficiency and effectiveness of the smart cities. Recently, the application of deep learning in smart cities has gained a tremendous attention from the research community. However, despite raise in popularity and achievements made by deep learning in solving problems in smart cities, no survey has been dedicated mainly on the application of deep learning in smart cities to show recent progress and direction for future development. To bridge this gap, this paper proposes to conduct a dedicated survey on the applications of deep learning in smart cities. In this paper, recent progress, new taxonomies, challenges and opportunities for future research opportunities on the application of deep learning in smart cities have been unveiled. The paper can provide opportunities for experts in the research community to propose a novel approach for developing the research area. On the other hand, new researchers interested in the research area can use the paper as an entry point. [ABSTRACT FROM AUTHOR]

The paper is a comprehensive review of the current research trends in the area of Arabic language especially state-of-the-art approaches to highlight the current status of diverse research aspects of that area to facilitate the adaption and extension of previous systems into new applications and systems. The Arabic language has deep, widespread and unexplored scope to research although the tremendous effort and researches that had been done previously. Modern state-of-the-art methods and approaches with fewer errors are required according to the high speed of hardware and technology development. The focus of this article will be on the offline Arabic handwritten text recognition as it is one of the most important topics in the Arabic scope. The main objective of this paper is critically analyzing the current researches to identify the problem areas and challenges faced by the previous researchers. This identification is intended to provide many recommendations for future advances in the area. It also compares and contrasts technical challenges, methods and the performances of handwritten text recognition previous researches works. It summarizes the critical problems and enumerates issues that should be considered when addressing these tasks. It also shows some of the Arabic datasets that can be used as inputs and benchmarks for training, testing and comparisons. Finally, it provides a fundamental comparison and discussion of some of the remaining open problems and trends in that field. [ABSTRACT FROM AUTHOR]

This paper aims to propose a new technique for identifying and categorizing malevolent Internet traffic within the context of security for smart devices. Given the rising usage of smart devices, including mobile phones, wearables, smart transportation-based devices, and the Internet of Things, concerns regarding their security are increasing. The need to develop effective security measures arises from the potential for attackers to compromise user data. In this study, we introduce an innovative approach that combines deep learning techniques, specifically convolutional neural networks (CNN), with long short-term memory (LSTM) for the purpose of detecting and categorizing malevolent Internet traffic. The objective of the proposed technique is to address the challenges related to time estimation by focusing on level prediction, resulting in a substantial reduction in prediction time for the identification of malevolent traffic. We utilize bidirectional long short-term memory–CNN (BI-LSTM–CNN) to identify malevolent communication and provide support for voice input. Experimental outcomes illustrate the effectiveness of our proposed technique in terms of precision, accuracy, F1 factor, false acceptance rate (FAR), false positive rate (FPR), and detection rate. In comparison with existing methods for detecting malevolent traffic, our approach achieves a 99.62% traffic detection rate, 99.98% accuracy, and 0.01% FAR, whereas the accuracy, detection rate of malevolent traffic, and FAR of existing methods are 99.88%, 97.32%, and 4.31%, respectively. These outcomes emphasize the superior performance and analysis of our technique, rendering it a valuable contribution to the realm of smart device security. In summary, this paper proposes a novel BI-LSTM–CNN technique for detecting malevolent traffic in smart devices. The proposed methodology tackles time estimation challenges and exhibits superior performance when compared to existing techniques. [ABSTRACT FROM AUTHOR]

The evolution of social media and online platforms created more opportunities to express one's thoughts without any burden. This emergence paved the way to construct more intelligent systems but by taking advantage of these platforms some people display their vengeance by harassing or abusing comments. The attackers leak sensitive information about an individual or an organization through the internet using abusive words, hate comments, threatening information, fake news, etc. So, online shaming detection with an optimum classification mechanism becomes necessary to avoid such negative consequences. In this paper, a novel online shaming detection mechanism using a Convolutional Neural Network Bidirectional Gated Recurrent Unit-based Hybrid Archimedes Tasmanian Devil (CNNBiGRU-HATD) approach is proposed to detect and classify the offensive comments and non-offensive comments. If the comment seems offensive, it is further classified into six categories according to shaming types such as abusive, doxing, Passing Judgment, Ethnic/Religious, Joke/Sarcasm, and Whataboutery. The utilized four Twitter-based datasets are initially preprocessed and feature extracted to improve the classification performance of the CNNBiGRU-HATD approach. After accurate classification, the information from the classified shaming types is collected and authenticated. The proposed approach is simulated using the Python platform. To validate the potential capability of the proposed CNNBiGRU-HATD approach, it is related to the existing methods by various parameters. The experimental result reveals that the proposed CNNBiGRU-HATD method achieves higher accuracy of 96.5% for the labeled hate speech detection dataset. [ABSTRACT FROM AUTHOR]

Emotion recognition (ER) from facial images is one of the landmark tasks in affective computing with major developments in the last decade. Initial efforts on ER relied on handcrafted features that were used to characterize facial images and then feed to standard predictive models. Recent methodologies comprise end-to-end trainable deep learning methods that simultaneously learn both, features and predictive model, where the most successful models are based on convolutional neural networks (CNNs). While these models have excelled at this task, they still fail at capturing local patterns that emerge in the learning process. We hypothesize these patterns could be captured by variants based on locally weighted learning. Specifically, in this paper we propose a CNN-based architecture enhanced with multiple branches formed by radial basis function (RBF) units that aims at exploiting local information at the final stage of the learning process (i.e., in the layers close to the output layer). Intuitively, these RBF units capture local patterns shared by similar instances using an intermediate representation, then the outputs of the RBFs are feed to a softmax layer that exploits this information to improve the predictive performance of the model. This feature could be particularly advantageous in ER as cultural / ethnicity differences may be potentially identified by the local units. We evaluate the proposed method in several ER datasets and show the proposed methodology achieves state-of-the-art performance in some of them, even when we adopt a pre-trained VGG-Face model as backbone. We show the proposed method outperforms consistently CNNs that do not have the proposed local learning component. Moreover, we found the proposed model is advantageous when training datasets are reduced and when merging images coming from different distributions (e.g., combining two ER datasets). We show it is the incorporation of local information what makes the proposed model competitive. [ABSTRACT FROM AUTHOR]

According to the World Health Organization and other authorities, falls are one of the main causes of accidental injuries among the elderly population. Therefore, it is essential to detect and predict the fall activities of older persons in indoor environments such as homes, nursing, senior residential centers, and care facilities. Due to non-contact and signal confidentiality characteristics, radar equipment is widely used in indoor care, detection, and rescue. This paper proposes an adaptive channel selection algorithm to separate the activity signals from the background using an ultra-wideband radar and to generalize fused features of frequency- and time-domain images which will be sent to a lightweight convolutional neural network to detect and recognize fall activities. The experimental results show that the method is able to distinguish three types of fall activities (i.e., stand to fall, bow to fall, and squat to fall) and obtain a high recognition accuracy up to 95.7%. [ABSTRACT FROM AUTHOR]

Automatic segmentation of optic disc (OD) and optic cup (OC) is an essential task for analysing colour fundus images. In clinical practice, accurate OD and OC segmentation assist ophthalmologists in diagnosing glaucoma. In this paper, we propose a unified convolutional neural network, named ResFPN-Net, which learns the boundary feature and the inner relation between OD and OC for automatic segmentation. The proposed ResFPN-Net is mainly composed of multi-scale feature extractor, multi-scale segmentation transition and attention pyramid architecture. The multi-scale feature extractor achieved the feature encoding of fundus images and captured the boundary representations. The multi-scale segmentation transition is employed to retain the features of different scales. Moreover, an attention pyramid architecture is proposed to learn rich representations and the mutual connection in the OD and OC. To verify the effectiveness of the proposed method, we conducted extensive experiments on two public datasets. On the Drishti-GS database, we achieved a Dice coefficient of 97.59%, 89.87%, the accuracy of 99.21%, 98.77%, and the Averaged Hausdorff distance of 0.099, 0.882 on the OD and OC segmentation, respectively. We achieved a Dice coefficient of 96.41%, 83.91%, the accuracy of 99.30%, 99.24%, and the Averaged Hausdorff distance of 0.166, 1.210 on the RIM-ONE database for OD and OC segmentation, respectively. Comprehensive results show that the proposed method outperforms other competitive OD and OC segmentation methods and appears more adaptable in cross-dataset scenarios. The introduced multi-scale loss function achieved significantly lower training loss and higher accuracy compared with other loss functions. Furthermore, the proposed method is further validated in OC to OD ratio calculation task and achieved the best MAE of 0.0499 and 0.0630 on the Drishti-GS and RIM-ONE datasets, respectively. Finally, we evaluated the effectiveness of the glaucoma screening on Drishti-GS and RIM-ONE datasets, achieving the AUC of 0.8947 and 0.7964. These results proved that the proposed ResFPN-Net is effective in analysing fundus images for glaucoma screening and can be applied in other relative biomedical image segmentation applications. [ABSTRACT FROM AUTHOR]

Label distribution learning achieved promising results on ordinal regression tasks such as facial age and attractiveness estimation, especially using deep label distribution learning (DLDL) methods, introducing the label distribution learning into deep convolutional neural networks. However, existing DLDL methods have an inconsistency between the training objectives and the evaluation metric, so they may be suboptimal. In addition, these methods always adopt image classification or face recognition models with a large amount of parameters, which carry expensive computation cost and storage overhead. In this paper, we firstly analyze the essential relationship between two state-of-the-art methods (ranking CNN and DLDL) and show that the ranking method is in fact learning label distribution implicitly. This result thus firstly unifies two existing popular state-of-the-art methods into the DLDL framework. Second, in order to alleviate the inconsistency and reduce resource consumption, we design a lightweight network architecture and propose a unified framework which can jointly learn label distribution and regress expectation value. The effectiveness of our approach has been demonstrated on typical ordinal regression tasks including facial age and attractiveness estimation. Our method achieves new state-of-the-art results using the single model with 36× fewer parameters and 3× faster inference speed. [ABSTRACT FROM AUTHOR]

Camouflaged object detection (COD) aims to identify and segment items that are seamlessly assimilate into the surroundings. Compared with the traditional image segmentation, the indefinable boundaries of camouflaged objects and high intrinsic similarities between the targets and the surrounding background make COD more challenging. Although many models have been designed to settle this issue, these algorithms suffer from inaccurate contours. Besides, existing COD methods mainly strive to expand the receptive field and excavate rich contextual cues using convolutional layers with large dilation rates. However, the tiny detailed information may be degenerated in this process due to the large "holes" in these dilated convolutional layers. This paper develops a novel boundary-guided context-aware network (BCNet) to deal with these problems. Specifically, we design a high-resolution feature enhancement module to excavate multi-scale information and enlarge the receptive field without destroying the tiny details within the input feature maps. The proposed module is effective in enhancing the single-layer feature and boosting the COD performance. Besides, a boundary-guided feature interaction module is designed to aggregate multi-level features and investigate the complementary relationship between the targets and the corresponding contours. We evaluate BCNet on four benchmark datasets to verify the effectiveness of the key modules. The experimental results demonstrate that BCNet achieves a real-time inference speed (49 FPS) on a Titan XP GPU and surpasses 19 cutting-edge contenders. Source code will be available at https://github.com/clelouch/BCNet. [ABSTRACT FROM AUTHOR]

Plant diseases are one of the major threats to the economy and food security of a country. Detection of such diseases timely and accurately on large scale is prone to human error. Techniques like machine learning (ML) and deep learning (DL) provide alternatives to build automated models that can detect such diseases efficiently. Several researchers have used deep learning techniques for plant disease detection. Fruit crops are major part of agricultural production. This paper proposes a lightweight and accurate deep learning model based on convolutional neural network (CNN) for the detection of diseased leaves in banana, guava and mango fruit crops. The model is proposed with the concept of feature reuse at three different levels. The model was trained using open database which consists of eight distinct classes of diseased and healthy leaves from three different fruit species. From the experiment, it was found that the model uses 101,000 numbers of parameters and achieves 99.14% success rate for disease leaves identification. Also, it outperforms 15 different state-of-the-art pre-trained models, namely VGG16, VGG19, ResNet50, ResNet50V2, ResNet152, ResNet152V2, InceptionV3, InceptionResNetV2, Xception, DenseNet121, DenseNet169, DenseNet201, MobileNetV2, ConvNeXtBase and ConvNeXtLarge, in terms of both accuracy and model complexity. [ABSTRACT FROM AUTHOR]

Traditional navigation systems rely mainly on satellite navigation, but internal positioning and navigation cannot be achieved. This is mainly due to the inability to penetrate the wall due to the complex internal environment and the signal reaching the ground. Therefore, this paper proposes the research of indoor positioning algorithm based on improved convolutional neural network. The main method of this paper is based on the monocular vision indoor positioning based on the improved convolutional neural network. This method is an intelligent solution to the problems of traditional methods such as high cost, poor anti-interference ability, weak robustness, and poor compatibility. Positioning aids. The traditional computer vision camera pose estimation method is affected by the complex background in the image. When extracting feature corners, it is very affected by non-interest corners, so this paper adds an improved convolutional neural network algorithm to add in complex indoor scenes. With the region limitation, camera pose estimation in the region of interest is better to achieve low-cost, high accuracy and more stability. The improved model in this paper was tested on a test set of indoor datasets with markers, and the recognition accuracy reached 98.1%. In addition, this paper improves the PnP solution method, compares and analyzes the traditional RPnP, EPnP, and CEPPnP algorithms, and seeks the optimal camera pose estimation algorithm to locate the world coordinates of the current shooting position. The experimental results show that this paper proposes the positioning algorithm has high stability. [ABSTRACT FROM AUTHOR]

Assessing the strength of complex networks is crucial for evaluating their functionality and monitoring system security. Two primary perspectives for measuring network robustness are connectivity and controllability, which quantify how well a network maintains connectivity and controllability after experiencing complete attacks. However, conventional robustness calculation methods that use simulation attacks can be impractical for large-scale networks. Recent research has explored the use of deep learning methods for network robustness prediction, including PCR (short for predictor of controllability robustness), which uses a VGG (short for visual geometry group)-based convolutional neural network (CNN) and a linear filter. However, PCR has limitations, including inaccurate edge importance differentiation and fluctuating output curves. This paper presents ATTRP, an attention-based robustness predictor, which addresses these limitations. ATTRP uses a novel convolutional neural network with channel and spatial attention mechanisms to improve prediction accuracy and a robustness filter based on Savitzky–Golay smoothing to rectify the robustness curve. Experimental studies demonstrate that ATTRP outperforms other predictors, with much lower errors, while its computational speed is far superior to that of traditional attacking simulation. Overall, our research provides a more effective and efficient means of predicting network robustness, with potential applications in monitoring system security. [ABSTRACT FROM AUTHOR]

Automated medical image segmentation plays an important role in many clinical applications, which however is a very challenging task, due to complex background texture, lack of clear boundary and significant shape and texture variation between images. Many researchers proposed an encoder–decoder architecture with skip connections to combine low-level feature maps from the encoder path with high-level feature maps from the decoder path for automatically segmenting medical images. The skip connections have been shown to be effective in recovering fine-grained details of the target objects and may facilitate the gradient back-propagation. However, not all the feature maps transmitted by those connections contribute positively to the network performance. In this paper, to adaptively select useful information to pass through those skip connections, we propose a novel 3D network with self-supervised function, named selective information passing network. We evaluate our proposed model on the MICCAI Prostate MR Image Segmentation 2012 Grant Challenge dataset, TCIA Pancreas CT-82 and MICCAI 2017 Liver Tumor Segmentation Challenge dataset. The experimental results across these datasets show that our model achieved improved segmentation results and outperformed other state-of-the-art methods. The source code of this work is available at https://github.com/ahukui/SIPNet. [ABSTRACT FROM AUTHOR]

Spectral efficiency (SE) optimization in massive multiple input multiple output (MIMO) antenna cognitive systems is a challenge originated from the coexistence restrictions. Traditional power allocation can optimize the SE; however, involving deep learning can meet real-time and fairness processing requirements. In unfair allocation problem, all power is possibly assigned to one or few antennas of a particular user. In this paper, we build a mathematical optimization model considering the fairness problem such that SE is optimized for all users. To implement the model, we propose an attention-based convolutional neural network (Att-CNN), where h 0 and h k (i.e., cross-interference and direct channels) attention mechanisms are used to improve the SE. The convolutional neural network is applied to decrease the floating point operations (FLOPs) and number of network parameters. We conducted experiments from these aspects: Fair antenna power allocation, power allocation performance and computational performance. Heat maps with different interference thresholds show the fair allocation for all users. Analyses of SE validate the superiority of the Att-CNN compared with the equal power allocation and fully connected neural network (FNN) schemes. The analyses of the FLOPs and number of parameters show the superiority of the Att-CNN over the FNN. [ABSTRACT FROM AUTHOR]

In recent years, there has been a renewal of interest in brain–computer interface (BCI). One of the BCI tasks is to classify the EEG motor imagery (MI). A great deal of effort has been made on MI classification. What seems to be lacking, however, is multiple MI classification. This paper develops a single-channel-based convolutional neural network to tackle multi-classification motor imagery tasks. For multi-classification, a single-channel learning strategy can extract effective information from each independent channel, making the information between adjacent channels not affect each other. A data evaluation method and a mutual information-based regularization parameters auto-selection algorithm are also proposed to generate effective spatial filters. The proposed method can be used to tackle the problem of an inaccurate mixed covariance matrix caused by fixed regularization parameters and invalid training data. To illustrate the merits of the proposed methods, we used the tenfold cross-validation accuracy and kappa as the evaluation measures to test two data sets. BCI4-2a and BCI3a data sets have four mental classes. For the BCI4-2a data set, the average accuracy is 79.01%, and the kappa is 0.7202 using data evaluation-based auto-selected filter bank regularized common spatial pattern voting (D-ACSP-V) and single-channel series convolutional neural network (SCS-CNN). Compared to traditional FBRCSP, the proposed method improved accuracy by 7.14% for the BCI4-2a data set. By using the BCI3a data set, the proposed method improved accuracy by 9.54% compared with traditional FBRCSP, the average accuracy of the proposed method is 83.70%, and the kappa is 0.7827. [ABSTRACT FROM AUTHOR]

Position detection is essential for precise contactless manipulation as it can play an important role in analyzing the behavior patterns and motion regularities of levitated objects. However, traditional detection methods have several limitations, such as finite feature representation, low detection accuracy, and poor adaptability, notably for small targets. To address these issues, this paper proposes an effective detector called RFRM-CenterNet, which is the first attempt to detect the location of levitated objects based on a convolutional neural network. First, a receptive field refinement module consisting of multi-stage parallel dilated convolutional layers is designed to detect small levitated objects. Then, to increase the feature representation capability, a feature fusion network is designed, which employs a parallel fusion of ResNet50 and a receptive field refinement module. In addition, an experimental system is constructed, and a dataset is established to train the model and verify the performance of the proposed method. The experimental results show that RFRM-CenterNet has better performance in detecting levitated objects than other detection methods. [ABSTRACT FROM AUTHOR]

The application of deep learning techniques in hyperspectral satellite images (HSI) classification has led to a significant increase in accuracy compared to machine learning techniques. The research objective is now shifted toward gaining classification accuracy with less number of spectral bands and training samples. With this research objective, the recent development of the hybrid convolution neural network (CNN) has resulted in the improvement of the accuracy of land use land cover classification of hyperspectral satellite images (HSI) and also provided a new dimension to explore. This paper proposes a variant of the 3D-2D CNN Hybrid model to extract representational features from a different class of land use land cover using different receptive fields in a multi-stack arrangement. The proposed model focused on reducing the number of bands required for classification and training samples but on the other hand improving the diverse representational feature extraction. The paper also explores through experimentation the possible replacement of monotonic activation function ReLu using non-monotonic functions like Swish and Mish in 3D-2D CNN Hybrid models. The experiments are carried out using publically available HSI datasets with a special focus on small training datasets to establish the efficiency and robustness of the proposed model. [ABSTRACT FROM AUTHOR]

To create a safe and secure cyberspace, many researchers have created and proposed Intrusion Detection Systems (IDSs). Different types of IDSs are either designed to address only a single kind of intrusion or multiple variants. In this paper, we proposed a new context-aware feature extraction method as a pre-processing step for Convolutional Neural Network (CNN)-based multiclass intrusion detection. Feature selection is also used to reduce the feature space and classification time. For attack-type classification, we used CNN which is generally well-known for its better classification ability on image recognition tasks. We have exploited this ability to develop an IDS that can identify different types of intrusions ranging from 4 to 12 types. In our study, we evaluated our proposed Context-aware Feature Extraction-based CNN IDS on different datasets with multiple classes where it was able to successfully improve classification accuracy compared to the models with no preprocessing steps and other existing methods. We carried out the performance evaluation and comparison using four different datasets, NSL-KDD, CICIDS2017, ADFA-LD, and ADFA-WD. These datasets have either host-based or network-based features, which is another area that we investigated in this paper. [ABSTRACT FROM AUTHOR]

The goal of steganalysis is to unearth a concealed message hidden inside an innocent carrier by steganography. Steganography in the hands of unlawful people can pose a great threat to society. Secret message is embedded as a noise (residue) which silently disrupts the statistical nature of the carrier without visual transformation. For the purpose of unearthing covert communication, modeling this noise (residue) is a crucial factor. In this paper, a hybrid deep learning framework is proposed with convolutional neural network (CNN) to emphasize the need for residual based investigation for digital image steganalysis. In order to model the noise residuals, five types of handcrafted features are extracted, in which the first four types rely on acquiring noise residuals based on filtering by linear and nonlinear filters, and the fifth type relies on residual acquisition method based on Empirical Mode Decomposition (EMD). The extracted features are categorized using a robust CNN with a new parallel architecture that is capable of learning intricate details from input features to classify it as cover or stego. Mostly CNNs are trained with raw images, but in the proposed method, the 1-dimensional residual features are stacked into 2-dimensional grid and are then used to train the CNN. Three spatial content-adaptive and four spatial non-content-adaptive algorithms are used to evaluate the performance of the proposed architecture. The experimental results show the versatility and robustness of the proposed hybrid architecture towards these algorithms when compared to existing state-of-the-art steganalytic methods. A practical issue prevalent in deploying steganalysis in real-world is 'mismatch scenario,' and based on the experimentation done in this paper, the proposed architecture performs well under stego algorithm mismatch and payload mismatch scenarios. [ABSTRACT FROM AUTHOR]

A dynamic texture refers to a sequence of images with regular and periodic changes in appearance from one image to another. Typical dynamic textures include burning flames, waving flags and flowing water. Given a reference dynamic texture, dynamic texture generation is meant to generate new dynamic textures that are of similar appearance to the reference one but are more diversified in terms of dynamics (i.e., periodic pattern changes). Existing dynamic texture generation algorithms may suffer from partially losing appearance information of the reference dynamic texture, resulting in such problems as structural distortion and blurred edges in the generated dynamic textures. To address this issue, we propose in this paper a structure-preserving dynamic texture generation algorithm based on a new two-stream convolutional networks model with feature reuse strategy (TSCN-FR). The proposed TSCN-FR makes use of spatial co-occurrence feature matrices to capture the spatial structure correlation of dynamic textures, thus alleviating the structural distortion and disordered color patch distribution in the generation of dynamic textures. In addition, an edge-preserving regular term, which uses the edge features of the reference dynamic texture extracted by the LoG operator, is introduced in the training of TSCN-FR to better maintain the edge structure of the reference dynamic texture. The application of our proposed method to dynamic style transferring is also presented. Experimental results have shown that our method for dynamic texture generation improves the existing methods by a clear margin on the ability to keep appearance information of the reference dynamic texture. [ABSTRACT FROM AUTHOR]

The detection of spatial and temporal changes (or change detection) in remote sensing images is essential in any decision support system about natural phenomena such as extreme weather conditions, climate change, and floods. In this paper, a new method is proposed to determine the inference process parameters of boundary point, rule coefficient, defuzzification coefficient, and dependency coefficient and present a new FWADAM+ method to train that set of parameters simultaneously. The initial data are clustered simultaneously according to each data group. This result will be the basis for determining a suitable set of parameters by using the FWADAM+ concurrent training algorithm. Eventually, these results will be inherited in the following data groups to build other complex fuzzy rule systems in a shorter time while still ensuring the model's efficiency. The weather imagery database of the United States Navy (US Navy) is used to evaluate and compare with some related methods using the root-mean-squared error (RMSE), R-squared (R2) measures, and the analysis of variance (ANOVA) model. The experimental results show that the proposed method is up to 30% better than the SeriesNet method, and the processing time is 10% less than that of the SeriesNet method. [ABSTRACT FROM AUTHOR]

The explosive growth of geospatial data is increasing requirements for automatic and efficient data learning abilities. Many deep learning methods have been widely applied for geospatial data understanding tasks, such as road networks and geospatial object detection. However, the demands for more accurate learning of high-level features require the use of deeper neural networks. To further improve the learning efficiency of deep neural networks, in this paper, we propose an improved convolutional neural network named EDense. First, we use its dense connectivity to integrate a CNN with an extreme learning machine. Then, we expand the kernels in the convolutional layers to increase the width of the network model. Furthermore, we propose one-feature EDense (OF-EDense), which is a simplified version of EDense, to fit conditions in which the number of parameters is strictly limited. Finally, the experimental results fully demonstrate the strong learning ability and high learning efficiency of EDense. [ABSTRACT FROM AUTHOR]

Convolutional neural network, a strong deep learning technique, is used to detect diseases and perform image processing, recognition, and disease classification. The neural network is a breakthrough in technology that can process large sets of images in both 2D and 3D. In this paper, a novel Deep CNN model named NPNet-19 is proposed to determine maize crop infections. The model's accuracy and robustness are tested using an enhanced dataset of 15,960 images to classify six disease classes and one healthy class. The primary data source for testing the model has been acquired from the maize fields situated in the state of Telangana. The model was trained using images from the Plant Village and Kaggle repositories. On training datasets, the learning model showed an accuracy of 97.51%, and on testing datasets, it achieved an accuracy of 88.72%. The proposed model is empirically compared with pre-trained models such as DenseNet-121, Inception V2, ShallowNet-8, and CNN-SVM, and it showed a 10.57, 1.74, 2.15, and 1.1% improvement in classification accuracy, respectively. When compared to transfer learning models including Modified LeNet, DICNN, SoyNet, Adaptive CNN, 9-layer, and 13-layered architectures, the proposed model showed a 10.12, 14.52, 8.17, 3.88, 7.25, and 3.39% improvement in classification accuracy, respectively. Classification Prediction Analysis also has been used for statistical analysis of prediction data on testing and training images using error metrics such as RMSE, MSE, and MAE. It is observed that the performance of the proposed model is better in the detection of real-time diseases and their classification. [ABSTRACT FROM AUTHOR]

Early diagnosis of brain tumor using magnetic resonance imaging (MRI) is vital for timely medication and effective treatment. But, most people living in remote areas do not have access to medical experts and diagnosis facilities. Nevertheless, recent advancement in the Internet of Thing and artificial intelligence is transforming the healthcare system and has led to the development of the Internet of Medical Things (IoMT). An automated brain tumor classification system integrated with the IoMT framework can aid in remotely diagnosing brain tumors. However, the existing methods for brain tumor classification in MRI based on traditional machine learning and deep learning are compute-intensive. Deployment of these methods in the real-world clinical setup poses a serious challenge. Therefore, there is a requirement for robust and compute-efficient techniques for brain tumor classification. To this end, this paper presents a novel lightweight attention-guided convolutional neural network (AG-CNN) for brain tumor classification in magnetic resonance (MR) images. The designed architecture uses channel-attention blocks to focus on relevant regions of the image for tumor classification. Besides, AG-CNN uses skip connections via global-average pooling to fuse features from different stages. This approach helps the network extract enhanced features essential to differentiate tumor and normal brain MR images. To access the efficacy of the designed neural network, we evaluated it on four benchmark brain tumor MRI datasets. The comparison results with the existing state-of-the-art methods revealed the robustness and computational efficiency of the proposed AG-CNN model. The designed brain tumor classification pipeline can be easily deployed on a resource-constrained embedded platform and used in real-world clinical settings to quickly classify brain tumors in MR images. [ABSTRACT FROM AUTHOR]

Ancient murals, suffering from severe diseases, usually exhibit the absence or distortion of local areas. The damaged murals severely impaired people's visual appreciation and satisfaction in the digital conservation of cultural heritage. However, there is no large amount of murals due to their scarcity. In this paper, we propose a novel content-constrained convolutional network for mural image completion. This method employs frequency transformation to facilitate effective multi-scale feature fusion for image inpainting, taking into account both space and frequency domains. Our network uses adaptive space-varying activation functions to correct feature maps across scales. Our network also uses dual-domain partial convolution with a mask for computing on only valid points, whereas the mask is updated for the next layer. This iterative process is performed until the mask is filled to build the repaired image. The proposed method is verified on the datasets in comparison with baseline methods. The experimental results demonstrate that the proposed method achieves better results with less artifacts in repairing mural images and generally outperforms the state-of-the-art methods both quantitatively and qualitatively. The code and pretrained models are available at https://github.com/zhangyongqin/C3N. [ABSTRACT FROM AUTHOR]

A convolutional neural network can be constructed using numerical methods for solving dynamical systems, since the forward pass of the network can be regarded as a trajectory of a dynamical system. However, existing models based on numerical solvers cannot avoid the iterations of implicit methods, which makes the models inefficient at inference time. In this paper, we reinterpret the pre-activation Residual Networks (ResNets) and their variants from the dynamical systems view. We consider that the iterations of implicit Runge–Kutta methods are fused into the training of these models. Moreover, we propose a novel approach to constructing network models based on high-order Runge–Kutta methods in order to achieve higher efficiency. Our proposed models are referred to as the Runge–Kutta Convolutional Neural Networks (RKCNNs). The RKCNNs are evaluated on multiple benchmark datasets. The experimental results show that RKCNNs are vastly superior to other dynamical system network models: they achieve higher accuracy with much fewer resources. They also expand the family of network models based on numerical methods for dynamical systems. [ABSTRACT FROM AUTHOR]

Automatic handwriting recognition is an important component for many applications in various fields. It is a challenging problem that has received a lot of attention in the past three decades. Research has focused on the recognition of Latin languages' handwriting. Fewer studies have been done for the Arabic language. In this paper, we present a new dataset of Arabic letters written exclusively by children aged 7–12 which we call Hijja. Our dataset contains 47,434 characters written by 591 participants. In addition, we propose an automatic handwriting recognition model based on convolutional neural networks (CNN). We train our model on Hijja, as well as the Arabic Handwritten Character Dataset (AHCD) dataset. Results show that our model's performance is promising, achieving accuracies of 97% and 88% on the AHCD dataset and the Hijja dataset, respectively, outperforming other models in the literature. [ABSTRACT FROM AUTHOR]

Computed tomography (CT) is a non-invasive diagnostic imaging modality that reveals more insight into human organs than conventional X-rays. In general, the CT output is a 3-D image that is formed by combining multiple 2D images or slices together. It is essential to keep in mind that not all of the slices provide significant information to detect tumours. Usually, a 3-D CT image obtained from the CT scanners has a significant number of unwanted non-organ slices in it. Radiologists typically devote a significant amount of time to select the slices with organ from a 3-D CT image. The presence of a tumour is only evident in the organ slice; hence, radiologists must be cautious not to skip any organ slices. This work is evaluated on the LITS, 3DIRCADb and COVID-19 CT datasets. The three datasets collectively contain 22,435 organ slices and 53,661 non-organ slices, and there is a huge gap between the number of organ and non-organ slices. There is a need for the automatic elimination of non-organ slices in 3-D CT volumes to assist the physicians, and hence, this work focuses on the automatic recognition of organ slices from 3-D CT volumes. In this paper, a new deep model called the computed tomography slice classification network (CTSC-Net) is proposed for CT slice classification between organ and non-organ slices. The model is trained on 77,980 CT slices, validated on 9748 slices and tested on 12,571 slices. Nine CNN architectures with different layer settings are trained and tested to arrive at the final optimal model. The performance measures are computed in terms of true positive rate, true negative rate, sensitivity, specificity and accuracy. The 20-layer CTSC-Net achieves a validation accuracy of 95.04% and an overall testing accuracy of 99.96%. The proposed model is compared to eight different pre-trained CNN models, and the results of the proposed CTSC-Net surpassed all the comparable models. The activation feature maps of different layers of the CTSC-Net are visualized to verify the discriminative features learned by the network. Hence, the proposed CTSC-Net can be employed as a computer-aided diagnosis tool to help physicians discard unnecessary non-organ slices from the 3-D CT volume and to speed up the CT diagnosis process. [ABSTRACT FROM AUTHOR]

Emotion is an instinctive or intuitive feeling as distinguished from reasoning or knowledge. It varies over time, since it is a natural instinctive state of mind deriving from one's circumstances, mood, or relationships with others. Since emotions vary over time, it is important to understand and analyze them appropriately. Existing works have mostly focused well on recognizing basic emotions from human faces. However, the emotion recognition from cartoon images has not been extensively covered. Therefore, in this paper, we present an integrated Deep Neural Network (DNN) approach that deals with recognizing emotions from cartoon images. Since state-of-works do not have large amount of data, we collected a dataset of size 8 K from two cartoon characters: 'Tom' & 'Jerry' with four different emotions, namely happy, sad, angry, and surprise. The proposed integrated DNN approach, trained on a large dataset consisting of animations for both the characters (Tom and Jerry), correctly identifies the character, segments their face masks, and recognizes the consequent emotions with an accuracy score of 0.96. The approach utilizes Mask R-CNN for character detection and state-of-the-art deep learning models, namely ResNet-50, MobileNetV2, InceptionV3, and VGG 16 for emotion classification. In our study, to classify emotions, VGG 16 outperforms others with an accuracy of 96% and F1 score of 0.85. The proposed integrated DNN outperforms the state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Sea fog is a dangerous weather phenomenon that seriously affects maritime traffic and other operations at sea. The conventional sea fog detection methods are not only difficult to make full advantage of the multispectral information of cloud images, but also deficient in the exploration of deep-level semantic information, leading to poor detection results. In this paper, we proposed a multimodal self-supervised convolutional neural network incorporating intra-modal band attention mechanism (MSCNN-IBAM) based on multispectral images of Himawari-8. MSCNN-IBAM uses independent branches to extract features from different modality cloud images and characterize the importance of each band through attention mechanisms. Simultaneously, multimodal self-supervised learning and supervised learning are effectively combined to optimize the model by constructing a two-tuple trainset. Experimental results show the accuracy, precision, recall, and F1 score of the proposed method as 97.72%, 95.84%, 96.54%, and 96.08%, respectively, which have the competitive performance and acceptable computational efficiency. And the additional analysis of sea fog cases shows that the proposed method is not only effective in identifying sea fog, but also has the ability to locate sea fog regions. [ABSTRACT FROM AUTHOR]

Diabetic Retinopathy (DR), one of the most progressive sight-threatening diseases caused by the long-term diabetic condition, can lead to vision impairment and blindness later. Early diagnosis and timely treatment help control and avert DR from its progression. However, manual grading is exceptionally challenging and arduous due to the complex anatomical features in the retina. Therefore, developing an automated diagnostic method for screening DR is obligatory. This paper proposes a deep learning-enabled optimized feature selection approach to classify the stage of DR severity in the fundus image. At first, a pre-processing phase eradicates the noise and improves the contrast in the retinal fundus image. Then, blood vessel segmentation is performed using the Coherence Enhancing Energy Based Regularized Level Set Evolution method in the green channel fundus image. Subsequently, the optic disk is segmented with Canny Anisotropic Diffusion filter and morphological transformations. Next, the candidate lesion region is detected using an Attention-based Fusion Network (AFU-Net). Then, shape and texture features are extracted, and then, the optimal subset of features is selected using the Improved Harris Hawk Optimization algorithm. Finally, a deep Convolutional Neural Network classifies the DR stages, and the model weight is updated using the same algorithm. The proposed method achieved superior performance in two benchmark public datasets compared with the existing state-of-the-art methods using F1-score, accuracy, sensitivity, and specificity measures. [ABSTRACT FROM AUTHOR]

Breast cancer is the second leading cause of death in women; therefore, effective early detection of this cancer can reduce its mortality rate. Breast cancer detection and classification in the early phases of development may allow for optimal therapy. Convolutional neural networks (CNNs) have enhanced tumor detection and classification efficiency in medical imaging compared to traditional approaches. This paper proposes a novel classification model for breast cancer diagnosis based on a hybridized CNN and an improved optimization algorithm, along with transfer learning, to help radiologists detect abnormalities efficiently. The marine predators algorithm (MPA) is the optimization algorithm we used, and we improve it using the opposition-based learning strategy to cope with the implied weaknesses of the original MPA. The improved marine predators algorithm (IMPA) is used to find the best values for the hyperparameters of the CNN architecture. The proposed method uses a pretrained CNN model called ResNet50 (residual network). This model is hybridized with the IMPA algorithm, resulting in an architecture called IMPA-ResNet50. Our evaluation is performed on two mammographic datasets, the mammographic image analysis society (MIAS) and curated breast imaging subset of DDSM (CBIS-DDSM) datasets. The proposed model was compared with other state-of-the-art approaches. The obtained results showed that the proposed model outperforms the compared state-of-the-art approaches, which are beneficial to classification performance, achieving 98.32% accuracy, 98.56% sensitivity, and 98.68% specificity on the CBIS-DDSM dataset and 98.88% accuracy, 97.61% sensitivity, and 98.40% specificity on the MIAS dataset. To evaluate the performance of IMPA in finding the optimal values for the hyperparameters of ResNet50 architecture, it compared to four other optimization algorithms including gravitational search algorithm (GSA), Harris hawks optimization (HHO), whale optimization algorithm (WOA), and the original MPA algorithm. The counterparts algorithms are also hybrid with the ResNet50 architecture produce models named GSA-ResNet50, HHO-ResNet50, WOA-ResNet50, and MPA-ResNet50, respectively. The results indicated that the proposed IMPA-ResNet50 is achieved a better performance than other counterparts. [ABSTRACT FROM AUTHOR]

Similarity searches on heterogeneous information networks (HINs) have attracted wide attention from both industrial and academic areas in recent years; for example, they have been used for friend detection in social networks and collaborator recommendation in coauthor networks. The structural information on the HIN can be captured by multiple metapaths, and people usually utilize metapaths to design methods for similarity search. The rich semantics in HINs are not only structural information but also content stored in nodes. However, the content similarity of nodes was usually not valued in the existing methods. Although some researchers have recently considered both types of information in machine learning-based methods for similarity search, they have used structure and content information separately. To address this issue by balancing the influence of structure and content information flexibly in the process of searching, we propose a double channel convolutional neural network model for top-k similarity search, which uses path instances as model inputs and generates structure and content embeddings for nodes based on different metapaths. We design an attention mechanism to enhance the differences in metapaths for each node. Another attention mechanism is used to combine the content and structure information of nodes. Finally, an importance evaluation function is designed to improve the accuracy and make the model more explainable. The experimental results show that our search algorithm can effectively support top-k similarity search in HINs and achieve higher performance than existing approaches. [ABSTRACT FROM AUTHOR]