Search

Scene classification and recognition have always been one of the most challenging tasks of scene understanding due to the inherent ambiguity in visual scenes. The core of scene classification and recognition tasks is scene representation. Deep learning advances in computer vision, especially deep CNNs, have significantly improved scene representation in the last decade. Deep convolutional features extracted from deep CNNs provide discriminative representations of the images and are widely used in various computer vision tasks, such as scene classification. Deep convolutional features capture the appearance characteristics of the image and the spatial information about different image regions. Meanwhile, the semantic and context information obtained from high-level concepts about scene images, such as objects and their relationships, can significantly contribute to identifying scene images. Therefore, in this paper, we divide visual scenes into two categories, object-based and layout-based. Object-based scenes are scenes that have scene-specific objects and, based on those objects, can be described and identified. In contrast, the layout-based scenes do not have scene-specific objects and are described and identified based on the appearance and layout of the image. This paper proposes a new neural network model for representing and classifying visual scenes, which we call G-CNN (GNN-CNN). The proposed model includes two modules, feature extraction and feature fusion, and the feature extraction module composes of visual and semantic branches. The visual branch is responsible for extracting deep CNN features from the image, and the semantic branch is responsible for extracting semantic GNN features from the scene graph corresponding to the image. The feature fusion module is a novel two-stream neural network that fuses the CNN and GNN feature vectors to produce a comprehensive representation of the scene image. Finally, a fully-connected classifier classified the obtained comprehensive feature vector into one of the pre-defined categories. The proposed model has been evaluated on three benchmark scene datasets, UIUC Sports, MIT67, and SUN397, and obtained classification accuracy of 99.91%, 96.01%, and 85.32%, respectively. In addition, a new dataset named Scene40, which has been introduced in our previous paper, is also used for further evaluation of the proposed method. The comparison results based on classification accuracy criteria show that the proposed model can outperform the best previous methods on three benchmark scene datasets. [ABSTRACT FROM AUTHOR]

With the establishment of the "double carbon" goal, various industries are actively exploring ways to reduce carbon emissions. Cloud data centers, represented by cloud computing, often have the problem of mismatch between load requests and resource supply, resulting in excessive carbon emissions. Based on this, this paper proposes a complete method for cloud computing carbon emission prediction. Firstly, the convolutional neural network and bidirectional long-term and short-term memory neural network (CNN-BiLSTM) combined model are used to predict the cloud computing load. The real-time prediction power is obtained by real-time prediction load of cloud computing, and then the carbon emission prediction is obtained by power calculation. Develop a dynamic server carbon emission prediction model, so that the server carbon emission can change with the change of CPU utilization, so as to achieve the purpose of low carbon emission reduction. In this paper, Google cluster data is used to predict the load. The experimental results show that the CNN-BiLSTM combined model has good prediction effect. Compared with the multi-layer feed forward neural network model (BP), long short-term memory network model (LSTM), bidirectional long short-term memory network model (BiLSTM), modal decomposition and convolution long time series neural network model (CEEMDAN-ConvLSTM), the MSE index decreased by 52 % , 50 % , 34 % and 45 % respectively. [ABSTRACT FROM AUTHOR]

Mental stress is a common problem that affects individuals all over the world. Stress reduces human functionality during routine work and may lead to severe health defects. Early detection of stress is important for preventing diseases and other negative health-related consequences of stress. Several neuroimaging techniques have been utilized to assess mental stress, however, due to its ease of use, robustness, and non-invasiveness, electroencephalography (EEG) is commonly used. This paper aims to fill a knowledge gap by reviewing the different EEG-related deep learning algorithms with a focus on Convolutional Neural Networks (CNNs) and Long Short-Term Memory networks (LSTMs) for the evaluation of mental stress. The review focuses on data representation, individual deep neural network model architectures, hybrid models, and results amongst others. The contributions of the paper address important issues such as data representation and model architectures. Out of all reviewed papers, 67% used CNN, 9% LSTM, and 24% hybrid models. Based on the reviewed literature, we found that dataset size and different representations contributed to the performance of the proposed networks. Raw EEG data produced classification accuracy around 62% while using spectral and topographical representation produced up to 88%. Nevertheless, the roles of generalizability across different deep learning models and individual differences remain key areas of inquiry. The review encourages the exploration of innovative avenues, such as EEG data image representations concurrently with graph convolutional neural networks (GCN), to mitigate the impact of inter-subject variability. This novel approach not only allows us to harmonize structural nuances within the data but also facilitates the integration of temporal dynamics, thereby enabling a more comprehensive assessment of mental stress levels. [ABSTRACT FROM AUTHOR]

Due to the complex structure of most frame structure, a large amount of sensor data needs to be processed for damage diagnosis, which increases the computational cost of diagnosis models and poses a serious challenge to their fast, accurate, and efficient damage diagnosis. In order to address this issue, this paper proposes a novel lightweight damage diagnosis method of frame structure for mobile devices based on convolutional neural networks. This method first uses mean filtering to process the vibration data collected by sensors, and then innovatively combines two convolutional neural network models, ShuffleNet and GhostNet, to form a new lightweight convolutional neural network model called SGNet, thereby reducing the computational cost of the model while ensuring diagnosis accuracy. In order to test the performance of the method proposed in this article, experimental research on damage degree diagnosis and damage type diagnosis is conducted by taking the frame structure provided by Columbia University as the research object, and comparative experiments of performance are conducted with MobileNet, GhostNet, and ShuffleNet. The experimental results show that the lightweight damage diagnosis method for frame structure proposed in this article not only has high damage diagnosis accuracy, but also has fewer computational parameters, when the highest accuracy is 99.8%, the computational parameters are only 1 million. At the same time, it is superior to MobileNet, GhostNet, ShuffleNet in terms of diagnosis accuracy and computational cost, so it is an effective high-precision lightweight damage diagnosis method for frame structure. [ABSTRACT FROM AUTHOR]

Protein structure prediction is one of the main research areas in the field of Bio-informatics. The importance of proteins in drug design attracts researchers for finding the accurate tertiary structure of the protein which is dependent on its secondary structure. In this paper, we focus on improving the accuracy of protein secondary structure prediction. To do so, a Multi-scale convolutional neural network with a Gated recurrent neural network (MCNN-GRNN) is proposed. The novel amino acid encoding method along with layered convolutional neural network and Gated recurrent neural network blocks helps to retrieve local and global relationships between features, which in turn effectively classify the input protein sequence into 3 and 8 states. We have evaluated our algorithm on CullPDB, CB513, PDB25, CASP10, CASP11, CASP12, CASP13, and CASP14 datasets. We have compared our algorithm with different state-of-the-art algorithms like DCNN-SS, DCRNN, MUFOLD-SS, DLBLS_SS, and CGAN-PSSP. The Q3 accuracy of the proposed algorithm is 82–87% and Q8 accuracy is 69–77% on different datasets. [ABSTRACT FROM AUTHOR]

Speech Emotion Recognition (SER) is a challenging task, and the typical convolutional neural network (CNN) cannot well handle the speech data directly. Because CNN tends to understand local information and ignores the overall characteristics. This paper proposes a Capsule Network with Two-Way Attention MechanismTWACapsNet for short) for the SER problem. TWACapsNet accepts the spatial and spectral features as inputs, and the convolutional layer and the capsule layer are deployed to process these two types of features in two ways separately. After that, two attention mechanisms are designed to enhance the information obtained from the spatial and spectral features. Finally, the results of these two ways are combined to form the final decision. The advantage of TWACapsNet is verified by experiments on multiple SER data sets, and experimental results show that the proposed method outperforms the widely-deployed neural network models on three typical SER data sets. Furthermore, the combination of the two ways contributes to the higher and more stable performance of TWACapsNet. [ABSTRACT FROM AUTHOR]

Deep neural networks are extensively used in text mining and Natural Language Processing is to enable computers to understand, analyze, and generate natural language data, such as text or speech, but semantic resources, such as taxonomies and ontologies, are not fully included in deep learning. In this paper, we use Deep Convolutional Neural Network (Deep CNN) to classify research papers using the Computer Science Ontology, an ontology of research areas in the field of computer science. It takes as input the abstract and keywords of a particular research paper and returns the relevant research topic. To evaluate our ontology, we used a gold standard dataset that includes research articles. To further improve text classification results, we propose to design a Deep CNN model. We then used ontology matching to reduce the classes and get better results. Experimental results show that the proposed approach outperforms the one with the highest precision, recall, and F1-score. [ABSTRACT FROM AUTHOR]

Accurate prediction of sea level height is critically important for the government in assessing sea level risk in coastal areas. However, due to the nonlinear, time-varying and highly uncertain characteristics of sea level change data, sea level prediction is challenging. To improve the accuracy of sea level prediction, this paper uses a new swarm intelligence algorithm named the sparrow search algorithm (SSA), which can imitate the foraging behavior and antipredation behavior of sparrows, to determine optimal solutions. To avoid the algorithm falling into a local optimal situation, this paper integrates the sine–cosine algorithm and the Cauchy variation strategy into the SSA to obtain an algorithm named the SCSSA. The SCSSA is used to optimize the parameter values of the CNN-BiLSTM (convolutional neural network combined with bidirectional long short-term memory neural network) model; finally, a combined neural network model (named SCSSA-CNN-BiLSTM) is proposed. In this paper, the time series data of seven tidal stations located in coastal China are used for experimental analysis. First, the SCSSA-CNN-BiLSTM model is compared with the CNN-BiLSTM model to predict the time series data of SHANWEI Station. With respect to the training and test sets of data, the SCSSA-CNN-BiLSTM model outperforms the other models on all the evaluation metrics. In addition, the remaining six tide station datasets and five neural network models, including the SCSSA-CNN-BiLSTM model, are used to further study the performance of the proposed prediction model. Four evaluation indices including the root mean squared error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE) and coefficient of determination (R2) are adopted. For six stations, the RMSE, MAE, MAPE and R2 of SCSSA-CNN-BiLSTM model are ranged from 20.9217 ~ 27.8427 mm, 9.4770 ~ 17.8603 mm, 0.1322% ~ 0.2482% and 0.9119 ~ 0.9759, respectively. The experimental analysis results show that the SCSSA-CNN-BiLSTM model makes effective predictions at all stations, and the prediction performance is better than that of the other models. Even though the combination of SCSSA algorithm may increase the complexity of the model, indeed the proposed model is a new prediction method with good accuracy and robustness for predicting sea level change. [ABSTRACT FROM AUTHOR]

Automatic handwritten character recognition plays a significant role in various applications across multiple fields. With the growing interest in automatic handwriting recognition and the advancement of deep learning methods, researchers have achieved significant improvements in the development of English handwriting recognition methods. However, the recognition of Arabic handwriting has received insufficient attention. In this paper, a novel "DeepAHR" model is presented to accurately and efficiently recognize Arabic handwritten characters using deep learning techniques. The "DeepAHR" model is based on a convolutional neural network (CNN) and is trained using two recent public datasets: Hijaa and Arabic handwritten characters dataset (AHCD). The overall accuracies of the proposed model were 98.66% and 88.24% on the AHCD and Hijaa datasets, respectively.The experimental results showed that DeepAHR outperformed state-of-the-art methods in the literature. These promising results provide evidence of the successful use of the DeepAHR model for recognizing handwritten Arabic characters [ABSTRACT FROM AUTHOR]

In recent years, vector-to-vector mapping-based raw waveform biosignal enhancement methods have gained significant attention in remote health monitoring system. In this paper, a novel end-to-end convolutional encoder–decoder (CED) model with stacked gated recurrent unit (SGRU) is proposed to learn sequential information of biosignal for signal enhancement. The proposed model CED-SGRU employs convolutional neural network to capture the spatial features and SGRU to capture temporal distributions of the biosignal layer by layer which increases the robustness of the proposed model. This work applies mean absolute error as a loss function for CED-SGRU-based vector-to-vector model. The proposed method has been evaluated on three foremost required biosignals, namely electrocardiogram, photoplethysmography and heart rate signals for diagnosing cardiovascular diseases. Experimental result shows outstanding denoising capability of the proposed CED-SGRU model on three biosignals which yields significantly higher reconstruction signal-to-noise ratio value and lower average absolute error, root mean square error and percent root mean square difference values when compared with state-of-the-art-methods. Moreover, the simple architecture of SGRU lowers the complexity of the model; thereby, reducing the inference time for denoising and restoring compressed biosignal tasks is fairly compared with baseline models, namely recurrent neural network model and long short-term memory model. [ABSTRACT FROM AUTHOR]

With the development of cloud computing, more and more resource constrained data owners tend to store their images in the cloud and rely on image retrieval services to get back images they want. However, this process raises huge privacy leakage risks. Privacy-preserving image retrieval (PPIR) can prevent privacy leakage during the retrieval process. Using convolutional neural networks for image feature extraction can significantly improve retrieval accuracy and has become a common trend in PPIR, but it also brings a new problem of network model parameter leakage, which is often ignored by most PPIR schemes. Besides, existing PPIR schemes often adopt centralized image retrieval service, which typically require frequent interactions between the cloud server and users during the retrieval process, resulting in excessive computational overhead for resource-limited users. Aiming to solve these problems, a distributed privacy-preserving image retrieval scheme is proposed in this paper. Three collaborative cloud servers utilize three neural network models split and outsourced by the data owner to extract features of the query image and perform subsequent retrieval tasks. A new secure Euclidean distance calculation protocol is designed for cloud servers to measure feature similarity, and secure sort pair protocol, secure sorting protocol, secure search protocol are utilized to achieve secure image retrieval. Additionally, a hierarchical index tree is employed to narrow down the search scope and improve search efficiency. This approach avoids the problem of network model parameter leakage caused by using neural networks to extract image features at the users' side, reduces the computational burden on users, and improves the accuracy and the efficiency of the retrieval process while ensuring retrieval security. Security analysis and experimental results demonstrate the security and retrieval performance of the proposed scheme. [ABSTRACT FROM AUTHOR]

Parkinson's disease (PD) is a condition that degenerates over time and impairs speech and pronunciation because brain cells have died. This research work aims to predict parkinson disease using the voice features extracted from speech signals recorded from PD individuals with dysphonic speech disorders by employing deep learning algorithms. PD is challenging to diagnose early on in the clinical presentation. To address the issue in machine learning methods, this paper proposes a neural network model by processing speech signals to classify PD using the University of California Irvine (UCI) machine learning repository dataset. Initially, a pre-loss reduction module is created by using pre-sampling to make the dataset balanced by reducing the dimensionality and maintaining the size of the space without influencing the learning process for data preparation. The relevant features are derived using a novel multi-agent salp swarm (MASS) algorithm, and a novel Parkinson classification neural network (PCNN) is proposed to classify Parkinson's patients with high accuracy employing these derived features. The result shows that the models that use MASS-PCNN produce higher classification accuracy of 99.1%, precision of 97.8%, recall of 94.7% and F1-score of 0.995 when paralleled to the existing models. As an outcome, the suggested model will perform superior to common convolutional neural networks. [ABSTRACT FROM AUTHOR]

Rolling element bearings are critical building blocks of any rotating machine. Achieving effective and precise fault diagnosis through various neural network models plays a pivotal role in ensuring the accuracy of rolling element bearing fault diagnosis. This research paper represented comparative study of artificial neural network (ANN), 1-D CNN, multi-input 1-D CNN, and 2-D CNN in fault diagnosis of rolling element bearings. The experiment was conducted on a roller bearing test rig over 2000 hours at constant speed of 800 rpm along with radial load of 1.5 kN till the development of naturally occurring operational surface defects on the bearing components. The proposed neural network architecture utilized multiple parallel convolutional layers to effectively extract rich and complementary fault features. The model was configured by implementing the categorical crossentropy loss function and Adam optimizer. Evaluation of the neural network models was performed using a confusion matrix and t-SNE visualization to ensure accurate fault identification. Comparative analysis among the convolutional neural network techniques was conducted to show their effectiveness toward fault diagnosis. The multi-input 1-D CNN achieved 97% prediction accuracy. The results demonstrate that multi-input 1-D CNN model provides better accuracy in fault diagnosis compared to the other models. [ABSTRACT FROM AUTHOR]

The health management of railway signal equipment in the high-speed railway is a key link between intelligent operation and maintenance. Accurately predicting the health state of compensation capacitors is of great significance to ensure the reliable work of track circuits. This paper proposes an improved deep neural network algorithm focusing on the problem of long-term accurate health forecasts for compensation capacitors. First, establishing a transmission state model for degradation mechanism mining, the difference function that can quantitatively evaluate features is defined by piecewise processing cab signaling receiving voltage. Introducing the degradation model, predictive driving under both model and data is implemented. Then, the convolutional neural networks and bidirectional long–short-term memory are combined and improved to construct a novel artificial intelligence combination strategy, while parameters are optimized based on the sparrow search algorithm. Finally, facing the conditional repair of compensation capacitors, we set a reasonable threshold for the occurrence of hidden dangers to complete fault warning. This novel and practical approach effectively explores the procedure of prognosis and health management, while the refined maintenance will better utilize current monitoring information, helping the intelligence and accuracy of safety control decision-making. [ABSTRACT FROM AUTHOR]

The prediction of wildfire spreading is necessary for managing and fighting the forest fire. The traditional models require higher accuracy of the input parameters, which is impossible in real forest fires. The paper proposed a fire-spreading model based on the dynamic data of the fire field to improve its adaptability. The model is designed using a convolutional neural network with mixed-inputs and attention mechanisms (MI-AM-CNN). It predicts the burn map after a period of time through the multiple-channel image containing terrain variables and the current burn map, and the scalars containing climate variables. The channel and spatial attention modules are integrated to handle the advanced features that contain important fire variables information and strengthen the influence of important features on the prediction. Based on the FARSITE, a large number of data sets are generated for training, validating, and testing the models in the paper. The proposed model MI-AM-CNN is compared with the state-of-the-art neural network models. Quantitative results show that MI-AM-CNN has the highest performance in predicting effectiveness and efficiency, and it can be applied recursively to get the continuous predicted results. In addition, the prediction results of MI-AM-CNN on the historical fire data demonstrate the ability of its application in the real fire case. The MI-AM-CNN can be used as a predictive method in firefighting operations, and its predicted results can provide theoretical support for the forest fire spread prediction method based on artificial intelligence. [ABSTRACT FROM AUTHOR]

With the increasing proportion of solar energy in the energy system, accurate solar irradiance forecast is of great significance for low-cost energy scheduling. This paper proposes a new forecasting idea for day-ahead solar irradiance forecast on the day-ahead scale: Firstly, based on formula derivation and big data correlation analysis, this paper finds out multiple parameters related to GHI, and jointly uses these parameters to forecast GHI. Error revision during morning period (ERDMP) is innovatively proposed on this basis, towards a more accurate daytime forecast. In order to prove the reliability and universality of the method, relevant data from five different-climatic regions are respectively used in the experiment. The multi-variable historical data-based day-ahead solar irradiance forecast uses deep neural networks, including Recurrent Neural Network (RNN), Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) network. ERDMP uses a linear AutoRegressive (AR) model to predict the daytime error coefficients based on the morning error coefficients. According to the results, through the proposed ERDMP, Mean Absolute Error (MAE) decreases by about 25% to 30%, Root Mean Squared Error (RMSE) decreases by about 20%, and R2 increases by about 5% to 10% when compared with the initial error of multi-parameter prediction models and other advanced models. [ABSTRACT FROM AUTHOR]

This paper delves into the specialized domain of human action recognition, focusing on the Identification of Indian classical dance poses, specifically Bharatanatyam. Within the dance context, a "Karana" embodies a synchronized and harmonious movement encompassing body, hands, and feet, as defined by the Natyashastra. The essence of Karana lies in the amalgamation of nritta hasta (hand movements), sthaana (body postures), and chaari (leg movements). Although numerous, Natyashastra codifies 108 karanas, showcased in the intricate stone carvings adorning the Nataraj temples of Chidambaram, where Lord Shiva's association with these movements is depicted. Automating pose identification in Bharatanatyam poses challenges due to the vast array of variations, encompassing hand and body postures, mudras (hand gestures), facial expressions, and head gestures. To simplify this intricate task, this research employs image processing and automation techniques. The proposed methodology comprises four stages: acquisition and pre-processing of images involving skeletonization and Data Augmentation techniques, feature extraction from images, classification of dance poses using a deep learning network-based convolution neural network model (InceptionResNetV2), and visualization of 3D models through mesh creation from point clouds. The use of advanced technologies, such as the MediaPipe library for body key point detection and deep learning networks, streamlines the identification process. Data augmentation, a pivotal step, expands small datasets, enhancing the model's accuracy. The convolution neural network model showcased its effectiveness in accurately recognizing intricate dance movements, paving the way for streamlined analysis and interpretation. This innovative approach not only simplifies the identification of Bharatanatyam poses but also sets a precedent for enhancing accessibility and efficiency for practitioners and researchers in the Indian classical dance. [ABSTRACT FROM AUTHOR]

Medical image fusion aims to fuse multiple images from a single or multiple imaging modes to enhance their corresponding clinical applications in diagnosing and evaluating medical problems, a trend that has attracted increasing attention. However, most recent medical image fusion methods require prior knowledge, making it difficult to select image features. In this paper, we propose a novel deep medical image fusion method based on a deep convolutional neural network (DCNN) for directly learning image features from original images. Specifically, source images are first decomposed by low rank representation to obtain the principal and salient components, respectively. Following that, the deep features are extracted from the decomposed principal components via DCNN and fused by a weighted-average rule. Then, considering the complementary between the salient components obtained by the low rank representation, a simple yet effective sum rule is designed to fuse the salient components. Finally, the fused result is obtained by reconstructing the principal and salient components. The experimental results demonstrate that the proposed method outperforms several state-of-the-art medical image fusion approaches in terms of both objective indices and visual quality. [ABSTRACT FROM AUTHOR]

This paper introduces a deep learning approach for accurately predicting the position and size of cracks in steam turbine blades using frequency response function data obtained from finite element analysis. Training deep neural networks for crack prediction necessitates an extensive dataset comprising various crack conditions, along with corresponding position and size information. However, obtaining sufficient high-quality and well-balanced experimental data for diverse crack conditions poses challenges. To address this issue, we leverage finite element analysis techniques to effectively amass a large volume of training data. Our methodology involves frequency response analysis, incorporating a crack modeling approach based on node sharing elimination and model order reduction (MOR). Automating this process enables us to efficiently generate a diverse dataset of crack blade frequency response data in a short timeframe, focusing on cracks in a single blade. The collected data is then transformed into two-dimensional image data, including magnitude similarity maps and magnitude frequency response maps, which serve as training data for our convolutional neural network (CNN) model. Evaluation of the CNN model's performance and accuracy for crack position and size prediction highlights its remarkable precision in determining the initiation position and size of blade cracks. Utilizing the proposed prediction model, there was a 95 % probability of accurately predicting the crack position within 3.1 % of the entire airfoil area. Furthermore, successful crack size predictions were achieved for approximately 96.4 % of the entire test dataset within an error range of ±1 mm. This underscores the potential capability of neural network training and crack prediction by leveraging training data derived from the finite element analysis and MOR. Our approach demonstrates promising implications for enhancing steam turbine blade maintenance and structural integrity assessment, as it streamlines data acquisition and provides an efficient solution for predicting critical crack characteristics. [ABSTRACT FROM AUTHOR]

This paper introduces the Hybrid Quantum Deep Neural Network (HQDNN), a pioneering model that amalgamates classical Convolutional Neural Network (CNN) architecture with Parameterized Quantum Circuits (PmQC) for image classification tasks. The background elucidates the intricacies of CNN and PmQC operations, emphasizing their respective components and functions. HQDNN comprises quantum convolutional and pooling layers alongside a classical fully connected layer, demonstrating its effectiveness through simulations on the MNIST dataset. Results reveal HQDNN's superior learning ability and classification performance compared to traditional CNN, particularly with the inclusion of a pooling layer. This work contributes to the evolution of quantum machine learning, opening avenues for the practical integration of quantum computing in image classification workflows. Future research may explore scalability and diverse applications of HQDNN, marking a promising convergence of quantum computing and deep learning methodologies. [ABSTRACT FROM AUTHOR]

A traditional approach for panoramic image generation consists of a random sample consensus (RANSAC) algorithm on a set of scale-invariant feature transform (SIFT) correspondences to generate a homography matrix between two images. Although producing adequate results for some type of images, hand-crafted SIFT features are not robust enough for highly varying natural images and the iterative RANSAC algorithm with its randomness does not always find the desired homography matrix. Recently, deep neural networks have been producing significant results in many challenging computer vision problems by learning features from large amounts of data. However, only very few recent works have been applied deep learning to panoramic image generation with the objective of finding feature correspondences and estimating homography matrix. Moreover, the absence of a proper dataset for the image stitching task hinders the standardization of models and comparison of their results. This paper attempts to generate panoramic images by extensively experimenting with various approaches using deep neural networks. The best proposed deep learning model achieved 7.31 and 1.07 pixels of the average absolute value loss for corner difference in X and Y directions, respectively. At the same time, qualitative results demonstrate superiority in comparison with the state-of-the-art SIFT+RANSAC algorithm. Specifically, in 72% of time the proposed framework either produced better results than SIFT+RANSAC or results of the proposed approach and SIFT+RANSAC were indistinguishable. Although SIFT + RANSAC produces better results in 28% of the time with respect to the loss function, our results are still visually comparable in many of these cases. Finally, a novel panoramic image generation dataset is introduced in this paper. [ABSTRACT FROM AUTHOR]

This paper investigates the fixed-time passivity of coupled quaternion-valued neural networks (CQVNNs) with multiple delayed couplings. Different from the finite-time passivity proposed in Wang et al. (IEEE Trans Cybern 50(5):2014–2025, 2020), a novel concept of fixed-time passivity is considered in this paper, where the settling time is independent on initial value. Compared with the existing literature, in this work we extend the passivity problem of CNNs from real field to quaternion field. In particular, the multiple coupling delays and discontinuous activation functions are considered, which have not been investigated in previous papers yet. By designing appropriate controllers and applying inequality technique, sufficient conditions are obtained correspondingly for three different types of fixed-time passivity of CQVNNs. Finally, simulations are given to substantiate the correctness of the theoretical results. [ABSTRACT FROM AUTHOR]

Near-infrared (NIR) imaging can add very useful data to many visible range image processing applications. In this paper, new fusion techniques are proposed to benefit from the data of both NIR/RGB sensors for the application of scene recognition and classification. Scene recognition and classification is an important and challenging branch of computer vision. Also, image fusion is a very well-known method in image processing. In this paper, fusion of RGB and NIR images is applied to improve the performance of scene classification. The proposed fusion technique is based on modified visual salient points (MVSP). Within this process, each RGB channel is fused with NIR channel based on their visual saliency maps. The fusion parameter space is then searched for the best fusion, based on an error function that is defined to measure the degree of conformity of the fused image with the input channels. This error function is defined on the basis of the distance between pixel intensities of NIR channel and fused image. On the contrary, the gradient difference is applied for the distance between RGB channels and fused image in the error function. After finding the most consent fused image, it is applied to deep convolutional neural networks (DCNNs) to perform scene category classification using the transfer learning method. The experimental results show that the proposed method in the scene classification outperforms the prior works. The results thus illustrate how an optimized NIR/RGB fusion can bring about better classification outcomes. [ABSTRACT FROM AUTHOR]

The convolutional neural network has been proven to be a robust recognition and diagnosis model for modelling diseases. In general, a convolutional neural network consists of feature learning which is considered a vital process in terms of feature extraction and classifier algorithms used for decision-making. In recent years, various pooling operations considered the spine of the feature learning layer have been developed in the literature; unfortunately, few of them can be spread to tackle the overfitting problem. In this paper, a novel pooling approach termed whole image average pooling is proposed to hold over the most significant feature information on the learning layer. The proposed method consists of two main stages: in the first stage, the average of all the values of the input image is calculated, in the second stage, the input image is divided into regions that represent the feature set of the input image, and then, the values of the regions are calculated. While obtaining the values of these features, the average of all values of the input image is taken into account. The proposed whole image average pooling approach addresses the overfitting problem by processing all pixel values in the image data and is used to get the important feature information from the medical datasets. To evaluate the performance of the proposed whole image average pooling, four clinically applicable tumour datasets as medical datasets have been trained and tested on seven convolutional neural network models AlexNet, VGG16, ResNet18, GoogleNet, ResNet50, R-CNN, and Faster R-CNN. The experimental outcomes demonstrate that the proposed whole image average pooling consistently outperformed the state-of-the-art pooling operation in terms of accuracy, sensitivity, specificity, positive predictive value, false-positive rate, and F1-measure evaluation criteria in three datasets. [ABSTRACT FROM AUTHOR]

The training of multilayer spiking neural networks (SNNs) using the error backpropagation algorithm has made significant progress in recent years. Among the various training schemes, the error backpropagation method that directly uses the firing time of neurons has attracted considerable attention because it can realize ideal temporal coding. This method uses time-to-first-spike (TTFS) coding, in which each neuron fires at most once, and this restriction on the number of firings enables information to be processed at a very low firing frequency. This low firing frequency increases the energy efficiency of information processing in SNNs. However, only an upper limit has been provided for TTFS-coded SNNs, and the information-processing capability of SNNs at lower firing frequencies has not been fully investigated. In this paper, we propose two spike-timing-based sparse-firing (SSR) regularization methods to further reduce the firing frequency of TTFS-coded SNNs. Both methods are characterized by the fact that they only require information about the firing timing and associated weights. The effects of these regularization methods were investigated on the MNIST, Fashion-MNIST, and CIFAR-10 datasets using multilayer perceptron networks and convolutional neural network structures. [ABSTRACT FROM AUTHOR]

Promoting the development of enterprises is a common task of all countries in the world, and the government has always taken promoting enterprise development as a long-term strategy. With the development for market, as an integral part of the social market economy, enterprises have made great contributions to the rapid of national economy. The enterprises are increasing, but the market competition is becoming increasingly fierce, and the weaknesses of enterprises such as limited scale, lax management and difficult financing are constantly highlighted. The appearance of these problems makes enterprises encounter many bottleneck problems in the process of development. Innovation, as the inexhaustible power of enterprise development, plays a significant role in enterprise growth, it is the key to solving the bottleneck problem of enterprise development. If an enterprise wants to achieve considerable development, it must innovate and enhance its competitiveness through innovation. In this context, how to evaluate enterprise innovation has become an important work. This work proposes an IPSO-ATT-MSCNN network via PSO from swarm optimization algorithm and artificial neural network to evaluate enterprise innovation. First, this work designs a multi-scale convolutional neural network (ATT-MSNN) via attention mechanism. This method uses multi-scale convolution to extract features of different scales, which improves the richness of features. This adds attention mechanisms to enhance useful features and reduce the impact of unwanted features such as noise. Second, in view of the performance degradation caused by the random initialization, this paper uses improved PSO algorithm to optimize the initial parameters. Third, this work proposes a series of strategies for promoting enterprise innovation. Finally, this work carried out a comprehensive and systematic experiment for the designed method, and the experiment verified the superiority of this method. [ABSTRACT FROM AUTHOR]

Forests play an essential role in ensuring ecological balance and providing valuable resources. Evaluating their ecological effectiveness and economic value is vital for sustainable forest management. Due to the ecological benefits of the forest ecosystem, China has introduced reforms in its ecological system and incorporated them into its national economic accounting system. This research study selects different types of forests based on their locality and species. It formulates a classifier using Convolution Neural networks (CNN) and Artificial Neural Networks (ANN) data mining techniques. First, this paper briefly discusses data mining methods used to determine the impact of different factors on forest health, quality, and its importance in terrestrial ecology. Second, this research study designs a classifier for labeling forests according to their ecological and economic worth. A confusion matrix was used to ascertain the performance of the classifier. Third, it illustrates an index system for evaluating forest ecosystem service function value and explores dynamic characteristics of forest resource variation and the change process of the forest ecosystem. Finally, a statistical analysis is performed among the proposed study and related studies. The proposed study outperforms its predecessors and depicts that the proposed technique is more accurate than traditional systems. The study showed that the amount of nutrients accumulated by trees and the amount of polluted gas absorbed have doubled, and the growth rate of other functions has also increased by more than 70%. The average growth of forest ecological functions is 97.93%, significantly improving people's awareness of forest resource protection. [ABSTRACT FROM AUTHOR]

Football (soccer) stands as the world's most popular sport, enthralling millions of fans globally. Within the dynamic sphere of modern football, gaining comprehensive insights into the nuanced technical and physical demands placed on players is pivotal for performance optimization. This paper introduces an innovative deep learning-powered image classification algorithm for recognizing football player activities directly from videos and images. Our pioneering approach harnesses the synergistic capabilities of convolutional neural networks (CNNs) and graph convolutional networks (GCNs) to decipher intricate spatial–temporal patterns in player poses and motions. The methodology employs a hybrid CNN and GCN architecture. The CNN leverages consecutive convolutional and pooling layers to automatically extract discriminative visual features from input frames capturing player poses. The GCN models' skeletal joints as graph nodes with bones as edges, performing graph convolutions to aggregate spatial and temporal information from neighboring body parts. This enables capturing localized pose dynamics. The complementary CNN and GCN outputs are fused through fully connected layers to classify player activities based on both visual appearances and pose configurations. The model is trained end to end on richly annotated football video data using a multi-class cross-entropy loss. Data augmentation and regularization techniques enhance robustness. Extensive experiments validate the proposed architecture's effectiveness, achieving 97.4% accuracy, 96% precision, 95.5% recall, 95.4% F1-score, 0.90 Matthews correlation, and 93% specificity in classifying 17 complex football activities significantly higher than previous benchmarks. Detailed ablation studies confirm the contributions of the CNN, GCN, and fused model components. The work represents a major advance in leveraging deep neural networks for accurate and granular analysis of sports performances. [ABSTRACT FROM AUTHOR]

The recent developments in unmanned aerial vehicles (UAVs) and indoor hybrid LiFi–WiFi networks (HLWNs) present a significant opportunity for creating low-cost, power-efficient, reliable, flexible, and ad-hoc HLWN-enabled indoor flying networks (IFNs). However, to efficiently operate and practically realize indoor HLWN, a unified physical layer (UniPHY) is indispensable for joint communication (control and data transfer) and sensing (e.g., localization). A UniPHY structure reduces costs and increases overall flexibility for HLWN-based IFNs. While conventional block-based wireless transceivers independently designed for LiFi and WiFi offer mediocre performance for a composite UniPHY waveform, a machine learning-based end-to-end learning framework for UniPHY can improve overall error performance and reduce the complexity of UAV transceiver hardware. Therefore, this paper proposes a novel generic end-to-end learning framework for a UniPHY system that can efficiently enable HLWN. The performance of the proposed learning framework based on deep neural networks (DNNs) and convolutional neural networks (CNNs) is investigated. Additionally, we assess the computational complexity of the proposed DNN and CNN learning frameworks. The results demonstrate that the performance of DNNs and CNNs varies depending on the considered channel model. Specifically, the analysis reveals that CNNs outperform traditional DNNs in WiFi (Rayleigh fading-based) channels. In contrast, traditional DNNs perform better than CNNs in LiFi (additive white Gaussian noise (AWGN)-based) channels. [ABSTRACT FROM AUTHOR]

Combining convolutional neural networks (CNNs) and recurrent neural networks (RNNs) produces a powerful architecture for video classification problems as spatial–temporal information can be processed simultaneously and effectively. Using transfer learning, this paper presents a comparative study to investigate how temporal information can be utilized to improve the performance of video classification when CNNs and RNNs are combined in various architectures. To enhance the performance of the identified architecture for effective combination of CNN and RNN, a novel action template-based keyframe extraction method is proposed by identifying the informative region of each frame and selecting keyframes based on the similarity between those regions. Extensive experiments on KTH and UCF-101 datasets with ConvLSTM-based video classifiers have been conducted. Experimental results are evaluated using one-way analysis of variance, which reveals the effectiveness of the proposed keyframe extraction method in the sense that it can significantly improve video classification accuracy. [ABSTRACT FROM AUTHOR]

The main objective of this paper is to improve the efficiency and accuracy of convolutional neural network models for image classification and denoising tasks. The focus of the study is on enhancing the activation layer of these models, which is a critical component that determines the output of each neuron in the network. To achieve this goal, we propose a novel activation function based on Chebyshev polynomials, which is both data-driven and self-learnable. In addition to proposing the LIP model, the authors investigate its performance in approximating various nonlinearities and determine its Lipschitz bound. The study then evaluates the performance of the proposed activation function by conducting experiments on multiple datasets using different convolutional neural network models. The results show that the proposed activation function outperforms other activation layers and significantly enhances the accuracy of image classification and denoising tasks. [ABSTRACT FROM AUTHOR]

Tea-cake content-based image retrieval (CBIR) is an essential issue in tea-cake traceability. Popular CBIR methods work on medical and social networking, whereas tea-cake images have intensive and parallel representations among classes. While facing tea-cake CBIR, low inter-class and high intra-class distances increase retrieval difficulties using traditional CBIR methods. Thus, this paper proposes a tea-cake CBIR approach based on deep neural networks to retrieve tea-cakes. In the model, we establish a feature extraction model with designed dense blocks, where a cross-entropy loss function is explored to train the model and catch detailed features. Furthermore, to decrease intra-class and expand inter-class interval, a masked autoregressive discriminative normalization flow (MADNF) is presented to map the gained features in high dimensions to corresponding representations in Gaussian spaces. Particularly, a maximum likelihood function is developed to train MADNF for avoiding non-convergence. Extensive experiments on the tea-cake dataset show our method has significant performance compared with current competitors. Furthermore, experiments on the bird species dataset further demonstrate the effectiveness of our proposed approach. [ABSTRACT FROM AUTHOR]

Nonconvex penalties are utilized for regularization in high-dimensional statistical learning algorithms primarily because they yield unbiased or nearly unbiased estimators for the parameters in the model. Nonconvex penalties existing in the literature such as SCAD, MCP, Laplace and arctan have a singularity at origin which makes them useful also for variable selection. However, in several high-dimensional frameworks such as deep learning, variable selection is less of a concern. In this paper, we present a nonconvex penalty which is smooth at origin. The paper includes asymptotic results for ordinary least squares estimators regularized with the new penalty function, showing asymptotic bias that vanishes exponentially fast. We also conducted simulations to better understand the finite sample properties and conducted an empirical study employing deep neural network architecture on three datasets and convolutional neural network on four datasets. The empirical study based on artificial neural networks showed better performance for the new regularization approach in five out of the seven datasets. [ABSTRACT FROM AUTHOR]

To reliably and accurately classify complicated "big" datasets, machine learning models must be continually improved. This research proposes straightforward yet competitive neural networks for text classification, even though graph neural networks (GNN) have reignited interest in graph-based text classification models. Convolutional neural networks (CNN), artificial neural networks (ANN), and their refined "fine-tuned" models (denoted as FT-CNN and FT-ANN) are the names given to our proposed models. The models presented in this paper demonstrate that our simple models like (CNN, ANN, FT-CNN, and FT-ANN) can perform better than more complex GNN ones such as (SGC, SSGC, and TextGCN) and are comparable to others (i.e., HyperGAT and Bert). The process of fine-tuning is also highly recommended because it improves the performance and reliability of models. The performance of our suggested models on five benchmark datasets (namely, Reuters (R8), R52, 20NewsGroup, Ohsumed, and Mr) is vividly illustrated. According to the experimental findings, on the majority of the target datasets, these models—especially those that have been fine-tuned—perform surprisingly better than SOTA approaches, including GNN-based models. [ABSTRACT FROM AUTHOR]

Fostered by technological and theoretical developments, deep neural networks (DNNs) have achieved great success in many applications, but their training via mini-batch stochastic gradient descent (SGD) can be very costly due to the possibly tens of millions of parameters to be optimized and the large amounts of training examples that must be processed. The computational cost is exacerbated by the inefficiency of the uniform sampling typically used by SGD to form the training mini-batches: since not all training examples are equally relevant for training, sampling these under a uniform distribution is far from optimal, making the case for the study of improved methods to train DNNs. A better strategy is to sample the training instances under a distribution where the probability of being selected is proportional to the relevance of each individual instance; one way to achieve this is through importance sampling (IS), which minimizes the gradients' variance w.r.t. the network parameters, consequently improving convergence. In this paper, an IS-based adaptive sampling method to improve the training of DNNs is introduced. This method exploits side information to construct the optimal sampling distribution and is dubbed regularized adaptive sampling (RAS). Experimental comparison using deep convolutional networks for classification of the MNIST and CIFAR-10 datasets shows that when compared against SGD and against another sampling method in the state of the art, RAS produces improvements in the speed and variance of the training process without incurring significant overhead or affecting the classification. [ABSTRACT FROM AUTHOR]

In this research paper, we study the capability of artificial neural network models to emulate storm surge based on the storm track/size/intensity history, leveraging a database of synthetic storm simulations. Traditionally, computational fluid dynamics (CFD) solvers are employed to numerically solve the storm surge governing equations that correspond to expensive to evaluate partial differential equations (PDE). This study presents a neural network model that can predict storm surge, informed by a database of synthetic storm simulations. This model can serve as a fast and affordable emulator for the expensive CFD solvers creating the original database. The neural network model is trained with the storm track parameters used to drive the CFD solvers, and the output of the model is the time-series evolution of the predicted storm surge across multiple nodes within the spatial domain of interest. Once the model is trained, it can be deployed for further predictions based on new storm track inputs. The developed neural network model is a time-series model, composed of a long short-term memory (LSTM), a variation of recurrent neural network (RNN), further enriched with convolutional neural networks (CNNs). The convolutional neural network is employed to capture the correlation of data spatially (across the aforementioned nodes). Therefore, the temporal and spatial correlations of data are captured by the combination of the mentioned models, representing the ConvLSTM model. As the problem is a sequence to sequence time-series problem, an encoder–decoder ConvLSTM model is designed. Furthermore, the performance of the developed convolutional recurrent neural network model is improved by residual connection networks. Additional techniques are employed in the process of model training to enrich the model performance that the model can learn from the data in a more effective way. The performance of the developed model is compared with the results provided by a Gaussian process (GP) implementation, representing a state-of-the-art alternative for establishing time-series emulation of storm surge predictions. The results show that the proposed convolutional recurrent neural network outperforms the GP implementation for the examined synthetic storm database. [ABSTRACT FROM AUTHOR]

Diabetic Retinopathy (DR) is the major cause of blindness, which seriously threatens the world's vision health. Limited medical resources make early diagnosis and a large-scale screening of DR difficult. Most of the current automatic diagnostic methods are mostly based on deep learning and large-scale labeled data. However, the insufficiency of manual annotations for medical images still is a great challenge of training deep neural networks. Self-supervised learning methods are proposed to learn general features from dataset without manual annotations. Inspired by this, we proposed a deep learning based DR classification model (SimCLR-DR). In this paper, we first use contrastive self-learning algorithm to pre-train the encoder based on convolution network with unlabeled retinal images, then retrain the encoder with classifier on a small annotated training data to detect referable DR. The experimental results on Kaggle dataset show that this proposed method can overcome the training data insufficiency problem and performs better than transfer learning. SimCLR-DR is a good beginning for other deep learning based medical image detection approaches facing the challenge of insufficient annotated data. Figure presents an overview of the proposed framework, which contains three main steps: (i) Data preprocessing; (ii) Pretext task of SimCLR-DR based on contrastive learning; (iii) Downstream Task of SimCLRDR based on CNN. [ABSTRACT FROM AUTHOR]

Deep Neural Networks (DNNs) are widely regarded as the most effective learning tool for dealing with large datasets, and they have been successfully used in thousands of applications in a variety of fields. Based on these large datasets, they are trained to learn the relationships between various variables. The adaptive moment estimation (Adam) algorithm, a highly efficient adaptive optimization algorithm, is widely used as a learning algorithm in various fields for training DNN models. However, it needs to improve its generalization performance, especially when training with large-scale datasets. Therefore, in this paper, we propose HN Adam, a modified version of the Adam Algorithm, to improve its accuracy and convergence speed. The HN_Adam algorithm is modified by automatically adjusting the step size of the parameter updates over the training epochs. This automatic adjustment is based on the norm value of the parameter update formula according to the gradient values obtained during the training epochs. Furthermore, a hybrid mechanism was created by combining the standard Adam algorithm and the AMSGrad algorithm. As a result of these changes, the HN_Adam algorithm, like the stochastic gradient descent (SGD) algorithm, has good generalization performance and achieves fast convergence like other adaptive algorithms. To test the proposed HN_Adam algorithm performance, it is evaluated to train a deep convolutional neural network (CNN) model that classifies images using two different standard datasets: MNIST and CIFAR-10. The algorithm results are compared to the basic Adam algorithm and the SGD algorithm, in addition to other five recent SGD adaptive algorithms. In most comparisons, the HN Adam algorithm outperforms the compared algorithms in terms of accuracy and convergence speed. AdaBelief is the most competitive of the compared algorithms. In terms of testing accuracy and convergence speed (represented by the consumed training time), the HN-Adam algorithm outperforms the AdaBelief algorithm by an improvement of 1.0% and 0.29% for the MNIST dataset, and 0.93% and 1.68% for the CIFAR-10 dataset, respectively. [ABSTRACT FROM AUTHOR]

The sentence-level sentiment classification is a classic topic of natural language processing, which aims to decide the sentiment tendency toward a sentence. However, previous studies ignore the significant role of words with sentimental tendencies in sentiment classification. In this paper, a sentiment information convolutional neural network (SI-CNN) model is proposed to break through this bottleneck problem. SI-CNN model contains three channels, where the first extracts original features from sentences, the second focuses on the words with sentiment tendencies, and the third is responsible for the categories and locations of the words with sentimental tendencies. We evaluate our model on three large-scale datasets. Experimental results show that the proposed SI-CNN outperforms other state-of-the-art deep neural networks and the introduction of sentiment information can improve the accuracy of sentiment classification. We also implement a series of exploratory experiments to prove the rationality of SI-CNN. [ABSTRACT FROM AUTHOR]

This document is an editorial discussing the development of 6G networks and the challenges and requirements they will face. The author emphasizes the need for research in areas such as autonomous systems, latency, vertical services, machine learning, trust and security, and quantum communications. The editorial also includes summaries of six research papers related to 6G networks, covering topics such as resource management, indoor positioning, resource allocation for connected cars, side-channel attacks, secure UAV communications, and reconfigurable intelligent surfaces. The authors of the editorial and the research papers come from various institutions and have expertise in communication and information systems, electronics and telecommunications engineering, electrical engineering, and computer science. [Extracted from the article]

Rapid and accurate detection of a C-shaped root canal on mandibular second molars can assist dentists in diagnosis and treatment. Oral panoramic radiography is one of the most effective methods of determining the root canal of teeth. There are already some traditional methods based on deep learning to learn the characteristics of C-shaped root canal tooth images. However, previous studies have shown that the accuracy of detecting the C-shaped root canal still needs to be improved. And it is not suitable for implementing these network structures with limited hardware resources. In this paper, a new lightweight convolutional neural network is designed, which combined with receptive field block (RFB) for optimizing feature extraction. In order to optimize the hardware resource requirements of the model, a lightweight, multi-branch, convolutional neural network model was developed in this study. To improve the feature extraction ability of the model for C-shaped root canal tooth images, RFB has been merged with this model. RFB has achieved excellent results in target detection and classification. In the multiscale receptive field block, some small convolution kernels are used to replace the large convolution kernels, which allows the model to extract detailed features and reduce the computational complexity. Finally, the accuracy and area under receiver operating characteristics curve (AUC) values of C-shaped root canals on the image data of our mandibular second molars were 0.9838 and 0.996, respectively. The results show that the deep learning model proposed in this paper is more accurate and has lower computational complexity than many other similar studies. In addition, score-weighted class activation maps (Score-CAM) were generated to localize the internal structure that contributed to the predictions. [ABSTRACT FROM AUTHOR]

Health monitoring systems are improving with the development of the internet of things. This paper proposes a secure architecture consisting of a four-layer internet of things enabled health monitoring system that collects patient data and classifies them into different medical categories. While collecting patient information from their wearable smart sensing devices for computation, the privacy and security of this process are essential. The main motive of this paper is to develop a lightweight and secure communication protocol using blockchain architecture for decentralized IoT networks and classify them into different categories using transfer learning. We propose a framework that uses blockchain for security and incorporates transfer learning to use multiple pre-trained models. The proposed routing technique uses factors like probability, credibility rating, and node energy to route the data to its destination such that the network overhead is reduced and the energy used is minimal. We classify the collected patient information using four different pre-trained convolutional neural network models: ResNet50, VGG19, InceptionV3, and SqueezeNet. We simulate the proposed routing approach and other benchmark schemes on various performance metrics. The results show that the proposed approach gives 92.24% classification accuracy. [ABSTRACT FROM AUTHOR]

Human facial analysis (HFA) has recently become an attractive topic for computer vision research due to technological progress and mobile applications. HFA explores several issues as gender recognition (GR), facial expression, age, and race recognition for automatically understanding social life. This study explores HFA from the angle of recognizing a person's gender from their face. Several hard challenges are provoked, such as illumination, occlusion, facial emotions, quality, and angle of capture by cameras, making gender recognition more difficult for machines. The Archimedes optimization algorithm (AOA) was recently designed as a metaheuristic-based population optimization method, inspired by the Archimedes theory's physical notion. Compared to other swarm algorithms in the realm of optimization, this method promotes a good balance between exploration and exploitation. The convergence area is increased By incorporating extra data into the solution, such as volume and density. Because of the preceding benefits of AOA and the fact that it has not been used to choose the best area of the face, we propose utilizing a wrapper feature selection technique, which is a real motivation in the field of computer vision and machine learning. The paper's primary purpose is to automatically determine the optimal face area using AOA to recognize the gender of a human person categorized by two classes (Men and women). In this paper, the facial image is divided into several subregions (blocks), where each area provides a vector of characteristics using one method from handcrafted techniques as the local binary pattern (LBP), histogram-oriented gradient (HOG), or gray-level co-occurrence matrix (GLCM). Two experiments assess the proposed method (AOA): The first employs two benchmarking datasets: the Georgia Tech Face dataset (GT) and the Brazilian FEI dataset. The second experiment represents a more challenging large dataset that uses Gallagher's uncontrolled dataset. The experimental results show the good performance of AOA compared to other recent and competitive optimizers for all datasets. In terms of accuracy, the AOA-based LBP outperforms the state-of-the-art deep convolutional neural network (CNN) with 96.08% for the Gallagher's dataset. [ABSTRACT FROM AUTHOR]

Hurricanes are tropical storms that cause immense damage to human life and property. Rapid assessment of damage caused by hurricanes is extremely important for the first responders. But this process is usually slow, expensive, labor intensive and prone to errors. The advancements in remote sensing and computer vision help in observing Earth at a different scale. In this paper, a new Convolutional Neural Network model has been designed with the help of satellite images captured from the areas affected by hurricanes. The model will be able to assess the damage by detecting damaged and undamaged buildings based upon which the relief aid can be provided to the affected people on an immediate basis. The model is composed of five convolutional layers, five pooling layers, one flattening layer, one dropout layer and two dense layers. Hurricane Harvey dataset consisting of 23,000 images of size 128 × 128 pixels has been used in this paper. The proposed model is simulated on 5750 test images at a learning rate of 0.00001 and 30 epochs with the Adam optimizer obtaining an accuracy of 0.95 and precision of 0.97. The proposed model will help the emergency responders to determine whether there has been damage or not due to the hurricane and also help those to provide relief aid to the affected people. [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 International Symposium on Equatorial Aeronomy (ISEA) is a conference that brings together researchers in atmospheric, ionospheric, and magnetospheric sciences to share new findings and discuss key questions in equatorial and low-latitude aeronomy. The 16th ISEA (ISEA-16) was held in September 2022 as a hybrid event, with both in-person and virtual participation. The symposium consisted of scientific sessions covering various topics, including equatorial irregularities, electrodynamics, space weather effects, and future trends in low-latitude aeronomy. The special issue of Earth, Planets & Space contains papers that summarize important scientific results presented at ISEA-16, covering topics such as equatorial plasma bubbles, volcanic eruptions' impacts on the upper atmosphere, machine learning techniques in aeronomy, and recent subjects in low-latitude aeronomy. The special issue contributes to advancing the field of equatorial aeronomy, and the next symposium, ISEA-17, will be held in Costa Rica in 2026. [Extracted from the article]

Solar energy-based technologies have developed rapidly in recent years, however, the inability to appropriately estimate solar energy resources is still a major drawback for these technologies. In this study, eight different artificial intelligence (AI) models namely; convolutional neural network (CNN), artificial neural network (ANN), long short-term memory recurrent model (LSTM), eXtreme gradient boost algorithm (XG Boost), multiple linear regression (MLR), polynomial regression (PLR), decision tree regression (DTR), and random forest regression (RFR) are designed and compared for solar irradiance prediction. Additionally, two hybrid deep neural network models (ANN-CNN and CNN-LSTM-ANN) are developed in this study for the same task. This study is novel as each of the AI models developed was used to estimate solar irradiance considering different timesteps (hourly, every minute, and daily average). Also, different solar irradiance datasets (from six countries in Africa) measured with various instruments were used to train/test the AI models. With the aim to check if there is a universal AI model for solar irradiance estimation in developing countries, the results of this study show that various AI models are suitable for different solar irradiance estimation tasks. However, XG boost has a consistently high performance for all the case studies and is the best model for 10 of the 13 case studies considered in this paper. The result of this study also shows that the prediction of hourly solar irradiance is more accurate for the models when compared to daily average and minutes timestep. The specific performance of each model for all the case studies is explicated in the paper. [ABSTRACT FROM AUTHOR]

Phononic metamaterials have the capability to manipulate the propagation of mechanical waves. The traditional finite element (FE) analysis-based methods for predicting phonon dispersion curves are computationally expensive for structure optimization that may require thousands of design evaluations, especially when applied to high-resolution metamaterial models with a large number of elements. To address this issue, this paper presents two physics-embedded deep convolutional neural networks to predict the phonon dispersion curves of 2D metamaterials: (1) a transfer learning-based convolutional neural network (TLCNN) and (2) a physics-guided convolutional neural network (PGCNN). The physics knowledge is embedded into the two proposed models by modifying the loss function of the convolutional neural network (CNN). A comparative study among CNN, TLCNN and PGCNN is conducted to understand the relative merits. The effectiveness of the physics-embedded methods is evaluated by comparing the predicted normalized eigenfrequencies with those obtained by direct numerical simulations (DNS) using FE simulation. Furthermore, a comparison of the computational costs, which include computing time and memory usage, is presented among DNS, CNN, TLCNN and PGCNN. It is demonstrated that the proposed TLCNN and PGCNN have the potential to improve prediction accuracy with a limited amount of input data. However, the computational costs of the "offline" model training are still significant. Among the three methods, PGCNN shows the best prediction accuracies on both the training and test sets. [ABSTRACT FROM AUTHOR]

Determining the rank of an elliptic curve E / Q is a difficult problem. In applications such as the search for curves of high rank, one often relies on heuristics to estimate the analytic rank (which is equal to the rank under the Birch and Swinnerton-Dyer conjecture). In this paper, we propose a novel rank classification method based on deep convolutional neural networks (CNNs). The method takes as input the conductor of E and a sequence of normalized Frobenius traces a p for primes p in a certain range ( p < 10 k for k = 3 , 4 , 5 ), and aims to predict the rank or detect curves of "high" rank. We compare our method with eight simple neural network models of the Mestre–Nagao sums, which are widely used heuristics for estimating the rank of elliptic curves. We evaluate our method on two datasets: the LMFDB and a custom dataset consisting of elliptic curves with trivial torsion, conductor up to 10 30 , and rank up to 10. Our experiments demonstrate that the CNNs outperform the Mestre–Nagao sums on the LMFDB dataset (remarkably, the neural network that took as an input all Mestre–Nagao sums performed much better than each sum individually). On the custom dataset, the performance of the CNNs and the Mestre–Nagao sums is comparable. [ABSTRACT FROM AUTHOR]

The visual saliency models based on low-level features of an image have the problem of low accuracy and scalability, while the visual saliency models based on deep neural networks can effectively improve the prediction performance, but require a large amount of training data, e.g., eye tracking data, to achieve good results. However, the traditional eye tracking method is limited by high equipment and time cost, complex operation process, low user experience, etc. Therefore, this paper proposed a visual saliency model based on crowdsourcing eye tracking data, which was collected by gaze recall with self-reporting from crowd workers. Parameter optimization on our crowdsourcing method was explored, and it came out that the accuracy of gaze data reached 1° of visual angle, which was 3.6% higher than other existed crowdsourcing methods. On this basis, we collected a webpage dataset of crowdsourcing gaze data and constructed a visual saliency model based on a fully convolutional neural network (FCN). The evaluation results showed that after trained by crowdsourcing gaze data, the model performed better, such as prediction accuracy increased by 44.8%. Also, our model outperformed the existing visual saliency models. We also applied our model to help webpage designers evaluate and revise their visual designs, and the experimental results showed that the revised design obtained improved ratings by 8.2% compared to the initial design. [ABSTRACT FROM AUTHOR]