Search

Visual information is important in final approach and landing phases for an approaching aircraft, it presents supplementary source for navigation system, and provides backup guidance when radio navigation fails, or even supports a complete vision‐based landing. Relative position and attitude can be solved from the runway features in the image. Traditional runway detection methods have high latency and low accuracy, which is unable to satisfy the requirements for a safe landing. This paper proposes a real‐time runway detection model, efficient runway feature extractor (ERFE), based on deep convolutional neural network, generating semantic segmentation and feature lines output. In order to evaluate the model's effectiveness, a benchmark is proposed to calculate the actual error between predicted feature line and ground truth one. A novel runway dataset which is based on pictures from Microsoft Flight Simulator 2020 (FS2020), is also proposed in this paper to train and test the model. The dataset will be released at https://www.kaggle.com/datasets/relufrank/fs2020‐runway‐dataset. ERFE shows excellent performance in FS2020 dataset, it gives satisfactory results even for real runway images excluded from our dataset. [ABSTRACT FROM AUTHOR]

Convolutional Neural Networks (CNNs) are a kind of artificial neural network designed to extract features and find out patterns for tasks such as segmentation, recognizing objects, and drawing up classification. Within a CNNs architecture, pooling operations are used until the number of parameters and the computational complexity are reduced. Numerous papers have focused on investigating the impact of pooling on the performance of Convolutional Neural Networks (CNNs), leading to the development of various pooling models. Recently, a fuzzy pooling operation based on type-1 fuzzy sets was introduced to cope with the local imprecision of the feature maps. However, in fuzzy set theory, it is not always accurate to assume that the degree of non-membership of an element in a fuzzy set is simply the complement of the degree of membership. This is due to the potential existence of a hesitation degree, which implies a certain level of uncertainty. To overcome this limitation, intuitionistic fuzzy sets (IFS) were introduced to incorporate the concept of a degree of hesitation. In this paper, we introduce a novel pooling operation based on intuitionistic fuzzy sets to incorporate the degree of hesitation heretofore neglected by a fuzzy pooling operation based on classical fuzzy sets, and we investigate its performance in the context of image classification. Intuitionistic pooling is performed in four steps: bifuzzification (by the transformation of data through the use of membership and non-membership maps), first aggregation (through the transformation of the IFS into a standard fuzzy set, second aggregation (through the transformation and use of a sum operator), and the defuzzification of feature map neighborhoods by using a max operator. IFS pooling is used for the construction of an intuitionistic pooling layer that can be applied as a drop-in replacement for the current, fuzzy (type-1) and crisp, pooling layers of CNN architectures. Various experiments involving multiple datasets demonstrate that an IFS-based pooling can enhance the classification performance of a CNN. A benchmarking study reveals that this significantly outperforms even the most recent pooling models, especially in stochastic environments. [ABSTRACT FROM AUTHOR]

In this paper, we discuss the implementation of artificial self-awareness mechanisms and self-reflection abilities in deep neural networks. While the current limitations of research prevent achieving cognitive capabilities on par with natural biological entities, the incorporation of basic self-awareness and self-reflection mechanisms in deep learning architectures offers substantial advantages in tackling specific problems across various scientific fields, including geosciences. In the first section, we outline the foundational architecture of our deep learning approach termed Self-Aware Learning (SAL). The subsequent part of the paper highlights the practical benefits of this machine learning methodology through synthetic tests and applications addressed to automatic classification and image analysis of real petrological data sets. We show how Self-Aware Learning allows enhanced accuracy, reduced overfitting problems, and improved performances compared to other existing methods. [ABSTRACT FROM AUTHOR]

The paramount concern of highly accurate energy-efficient computing in machines with significant cognitive capabilities aims to enhance the accuracy and efficiency of bio-inspired Spiking Neural Networks (SNNs). This paper addresses this main objective by introducing a novel spatial power spike-timing-dependent plasticity (Spatial-Pow-STDP) learning rule as a digital block with high accuracy in a bio-inspired SNN model. Motivated by the demand for precise and accelerated computation that reduces high-cost resources in neural network applications, this paper presents a methodology based on COordinate Rotation DIgital Computer (CORDIC) definitions. The proposed designs of CORDIC algorithms for exponential (Exp CORDIC), natural logarithm (Ln CORDIC), and arbitrary power function (Pow CORDIC) are meticulously detailed and evaluated to ensure optimal acceleration and accuracy, which respectively show average errors near 10–9, 10–6, and 10–5 with 4, 4, and 6 iterations. The engineered architectures for the Exp, Ln, and Pow CORDIC implementations are illustrated and assessed, showcasing the efficiency achieved through high frequency, leading to the introduction of a Spatial-Pow-STDP learning block design based on Pow CORDIC that facilitates efficient and accurate hardware computation with 6.93 × 10–3 average error with 9 iterations. The proposed learning mechanism integrates this structure into a large-scale spatiotemporal SNN consisting of three layers with reduced hyper-parameters, enabling unsupervised training in an event-based paradigm using excitatory and inhibitory synapses. As a result, the application of the developed methodology and equations in the computational SNN model for image classification reveals superior accuracy and convergence speed compared to existing spiking networks by achieving up to 97.5%, 97.6%, 93.4%, and 93% accuracy, respectively, when trained on the MNIST, EMNIST digits, EMNIST letters, and CIFAR10 datasets with 6, 2, 2, and 6 training epochs. [ABSTRACT FROM AUTHOR]

This paper discusses the problem of recognizing defective epoxy drop images for the purpose of performing vision-based die attachment inspection in integrated circuit (IC) manufacturing based on deep neural networks. Two supervised and two unsupervised recognition models are considered. The supervised models examined are an autoencoder (AE) network together with a multi-layer perceptron network (MLP) and a VGG16 network, while the unsupervised models examined are an autoencoder (AE) network together with k-means clustering and a VGG16 network together with k-means clustering. Since in practice very few defective epoxy drop images are available on an actual IC production line, the emphasis in this paper is placed on the impact of data augmentation on the recognition outcome. The data augmentation is achieved by generating synthesized defective epoxy drop images via our previously developed enhanced loss function CycleGAN generative network. The experimental results indicate that when using data augmentation, the supervised and unsupervised models of VGG16 generate perfect or near perfect accuracies for recognition of defective epoxy drop images for the dataset examined. More specifically, for the supervised models of AE+MLP and VGG16, the recognition accuracy is improved by 47% and 1%, respectively, and for the unsupervised models of AE+Kmeans and VGG+Kmeans, the recognition accuracy is improved by 37% and 15%, respectively, due to the data augmentation. [ABSTRACT FROM AUTHOR]

In recent years, the use of deep neural network in effective network feature extraction and the design of efficient and high-precision hyperspectral image classification algorithms has gradually become a research hotspot for scholars. However, due to the difficulty of obtaining hyperspectral images and the high cost of annotation, the training samples are very limited. In order to cope with the small sample problem, researchers often deepen the network model and use the attention mechanism to extract features; however, as the network model continues to deepen, the gradient disappears, the feature extraction ability is insufficient, and the computational cost is high. Therefore, how to make full use of the spectral and spatial information in limited samples has gradually become a difficult problem. In order to cope with such problems, this paper proposes two-branch multiscale spatial–spectral feature aggregation with a self-attention mechanism for a hyperspectral image classification model (FHDANet); the model constructs a dense two-branch pyramid structure, which can achieve the high efficiency extraction of joint spatial–spectral feature information and spectral feature information, reduce feature loss to a large extent, and strengthen the model's ability to extract contextual information. A channel–space attention module, ECBAM, is proposed, which greatly improves the extraction ability of the model for salient features, and a spatial information extraction module based on the deep feature fusion strategy HLDFF is proposed, which fully strengthens feature reusability and mitigates the feature loss problem brought about by the deepening of the model. Compared with five hyperspectral image classification algorithms, SVM, SSRN, A2S2K-ResNet, HyBridSN, SSDGL, RSSGL and LANet, this method significantly improves the classification performance on four representative datasets. Experiments have demonstrated that FHDANet can better extract and utilise the spatial and spectral information in hyperspectral images with excellent classification performance under small sample conditions. [ABSTRACT FROM AUTHOR]

A variant of membrane computing models called Spiking Neural P systems (SNP systems) closely mimics the structure and behavior of biological neurons. As third-generation neural networks, SNP systems have flexible architectures allowing the design of bio-inspired machine learning algorithms. This paper proposes Modular Spiking Neural P (MSNP) systems to solve image classification problems, a novel SNP system to be applied in scenarios where hundreds or even thousands of different classes are considered. A main issue to face in such situations is related to the structural complexity of the network. MSNP systems devised in this work allow to approach the general classification problem by dividing it in smaller parts, that are then faced by single entities of the network. As a benchmark dataset, the Oxford Flowers 102 dataset is considered, consisting of more than 8000 pictures of flowers belonging to the 102 species commonly found in the UK. These classes sometimes present large variations within them, may be also very similar to one another, and different images of the same subject may differ a lot. The work describes the architecture of the MSNP system, based on modules focusing on a specific class, their training phase, and the evaluation of the model both concerning result accuracy as well as energy consumption. Experimental results on image classification problems show that the model achieves good results, but is strongly connected to image quality, mainly depending on the frequency of images, remarkable changes of pose, images not centered, and subject mostly not shown. [ABSTRACT FROM AUTHOR]

A neural network model needs to be manually designed for ancient mural dynasty recognition, and this paper proposes an ancient mural dynasty recognition algorithm that is based on a neural architecture search (NAS). First, the structural edge information of mural images is extracted for use by the neural network model in recognizing mural missions. Second, an NAS algorithm that is based on contrast selection (CS) simplifies the architecture search to an incremental CS and then searches for the optimal network architecture on the mural dataset. Finally, the identified optimal network architecture is used for training and testing to complete the mural dynasty recognition task. The results show that the top accuracy of the proposed method on the mural dataset is 88.10%, the recall rate is 87.52%, and the precision rate is 87.69%. Each evaluation index used by the neural network model is superior to that of classical network models such as AlexNet and ResNet-50. Compared with NAS methods such as ASNG and MIGO, the accuracy of mural dynasty recognition is higher by an average of 4.27% when using the proposed method. The proposed method is verified on CIFAR-10, CIFAR-100, ImageNet16-120 and other datasets and achieves a good recognition accuracy in the NAS-bench-201 search space, which averages 93.26%, 70.73% and 45.34%, respectively, on the abovementioned datasets. [ABSTRACT FROM AUTHOR]

Deep neural network has made surprising achievements in natural language processing, image pattern classification recognition, and other domains in the last few years. It is still tough to apply to hardware-constrained or mobile equipment because of the huge number of parameters, high storage as well as computing costs. In this paper, a new sparse iteration neural network architecture is proposed. First, the pruning method is used to compress the model size and make the network sparse. Then the architecture is iterated on the sparse network model, and the network performance is improved without adding additional parameters. Finally, the hybrid deep learning model was carried out on CV tasks and NLP tasks on ANN, CNN, and Transformer. Compared with the sparse network architecture, we finally found that the accuracy of the MINST, CIFAR10, PASCAL VOC 2012, and SQuAD datasets is improved by 0.47%, 0.64%, 3.75%, and 15.06%, respectively. [ABSTRACT FROM AUTHOR]

Learning in neural networks with locally-tuned neuron models such as radial Basis Function (RBF) networks is often seen as instable, in particular when multi-layered architectures are used. Furthermore, universal approximation theorems for single-layered RBF networks are very well established; therefore, deeper architectures are theoretically not required. Consequently, RBFs are mostly used in a single-layered manner. However, deep neural networks have proven their effectiveness on many different tasks. In this paper, we show that deeper RBF architectures with multiple radial basis function layers can be designed together with efficient learning schemes. We introduce an initialization scheme for deep RBF networks based on k-means clustering and covariance estimation. We further show how to make use of convolutions to speed up the calculation of the Mahalanobis distance in a partially connected way, which is similar to the convolutional neural networks (CNNs). Finally, we evaluate our approach on image classification as well as speech emotion recognition tasks. Our results show that deep RBF networks perform very well, with comparable results to other deep neural network types, such as CNNs. [ABSTRACT FROM AUTHOR]

Intelligent rice disease recognition methods based on deep neural networks can predict the degree of disease on the basis of, for example, the number of disease spots on an image, so that preventive measures can be taken. Currently, intelligent recognition methods for rice diseases suffer from the disadvantages of poor versatility and low accuracy. This paper uses eight common image classification networks to classify and identify four rice diseases. ResNet50 was selected as the feature extraction network and an enhanced feature fusion and target adaptive network (EFFTAN), referred to as EFFTAN, is proposed. The EFFTAN was used to detect four rice spot diseases in the rice leaf disease image samples dataset; the mean average precision of the final detection was 95.3%, and effective detection was also achieved for the dense spot features. [ABSTRACT FROM AUTHOR]

The recognition of food images is of great significance for nutrition monitoring, food retrieval and food recommendation. However, the accuracy of recognition had not been high enough due to the complex background of food images and the characteristics of small inter-class differences and large intra-class differences. To solve these problems, this paper proposed a food image recognition method based on transfer learning and ensemble learning. Firstly, generic image features were extracted by using the convolutional neural network models (VGG19, ResNet50, MobileNet V2, AlexNet) pre-trained on the ImageNet dataset. Secondly, the 4 pre-trained models were transferred to the food image dataset for model fine-tuning. Finally, different basic learner combination strategies were adopted to establish the ensemble model and classify feature information. In this paper, several kinds of experiments were performed to compare the results of food image recognition between single models and ensemble models on food-11 dataset. The experimental results demonstrated that the accuracy of the ensemble model was the highest, reaching 96.88%, which was superior to any base learner. Therefore, the convolutional neural network model based on transfer learning and ensemble learning has strong learning ability and generalization ability, and it is feasible and practical to apply the method to food image recognition. [ABSTRACT FROM AUTHOR]

Large deep learning models are impressive, but they struggle when real-time data is not available. Few-shot class-incremental learning (FSCIL) poses a significant challenge for deep neural networks to learn new tasks from just a few labeled samples without forgetting the previously learned ones. This setup can easily leads to catastrophic forgetting and overfitting problems, severely affecting model performance. Studying FSCIL helps overcome deep learning model limitations on data volume and acquisition time, while improving practicality and adaptability of machine learning models. This paper provides a comprehensive survey on FSCIL. Unlike previous surveys, we aim to synthesize few-shot learning and incremental learning, focusing on introducing FSCIL from two perspectives, while reviewing over 30 theoretical research studies and more than 20 applied research studies. From the theoretical perspective, we provide a novel categorization approach that divides the field into five subcategories, including traditional machine learning methods, meta learning-based methods, feature and feature space-based methods, replay-based methods, and dynamic network structure-based methods. We also evaluate the performance of recent theoretical research on benchmark datasets of FSCIL. From the application perspective, FSCIL has achieved impressive achievements in various fields of computer vision such as image classification, object detection, and image segmentation, as well as in natural language processing and graph. We summarize the important applications. Finally, we point out potential future research directions, including applications, problem setups, and theory development. Overall, this paper offers a comprehensive analysis of the latest advances in FSCIL from a methodological, performance, and application perspective. • In-depth survey on FSCIL methods, applications, performance. • Categorizing FSCIL into five subcategories for clear analysis. • Evaluating FSCIL research on benchmarks for strengths, weaknesses. • FSCIL applications in computer vision, NLP, and graph. [ABSTRACT FROM AUTHOR]

For the past few years, sweat analysis for health monitoring has attracted increasing attention benefiting from wearable technology. In related research, the sensitive detection of uric acid (UA) in sweat with complex composition based on surface-enhanced Raman spectroscopy (SERS) for the diagnosis of gout is still a significant challenge. Herein, we report a visualized and intelligent wearable sweat platform for SERS detection of UA in sweat. In this wearable platform, the spiral channel consisted of colorimetric paper with Ag nanowires (AgNWs) that could capture sweat for SERS measurement. With the help of photos from a smartphone, the pH value and volume of sweat could be quantified intelligently based on the image recognition technique. To diagnose gout, SERS spectra of human sweat with UA are collected in this wearable intelligent platform and analyzed by artificial intelligence (AI) algorithms. The results indicate that the artificial neural network (ANN) algorithm exhibits good identification of gout with high accuracy at 97%. Our work demonstrates that SERS-AI in a wearable intelligent sweat platform could be a feasible strategy for diagnosis of gout, which expands research on sweat analysis for comfortable and noninvasive health monitoring. [ABSTRACT FROM AUTHOR]

Adversarial examples are often used to test and evaluate the security and robustness of image classification models. Though adversarial attacks under white-box setting can achieve a high attack success rate, due to overfitting, the success rate of black-box attacks is relatively low. To this end, this paper proposes diversified input strategies to improve the transferability of adversarial examples. In this method, various transformation methods are applied to randomly transform the original image multiple times, thereby generating a batch of transformed images. Then, in the process of back-propagation, the loss function gradient of the transformed images is calculated, and a weighted average of the obtained gradient values is performed to generate adversarial perturbation, which is iteratively added to the original image to generate adversarial examples. Meanwhile, by increasing the variety of data augmentation transformation types and the number of input images, the proposed method effectively alleviates overfitting and improves the transferability of adversarial examples. Extensive experiments on the ImageNet dataset indicate that the proposed method can perform black-box attacks better than benchmark methods, with an average of 97.2% success rate attacking multiple models simultaneously. [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]

The recognition of steel microstructure images plays a crucial role in the metallographic analysis process. Although some progress has been made through the application of artificial intelligence algorithms, several challenges remain. First, existing algorithms exhibit weak nonlinear feature extraction capabilities and noticeable limitations. Second, they overlook the intrinsic noise and redundant interference present in microscopic images. To address these issues, this paper investigates the automatic recognition of metallographic tissues by leveraging residual structures in deep neural networks. An enhanced residual network model based on transfer learning is proposed, which utilizes the pre-trained weights from the ImageNet dataset to facilitate learning with small sample data. This network offers higher classification accuracy and higher F1 scores. In addition, a deep residual shrinkage network model based on an attention mechanism is proposed. This model incorporates an attention sub-network into the original residual module and employs a soft threshold function to eliminate redundant features, including noise. The proposed algorithms are evaluated against various convolutional neural networks using 20 types of metallographic test sets. The experimental results showed that both methods have high accuracy rates of 95% and 94.44%, respectively, and F1 scores of 0.9464 and 0.9419. While maintaining the complexity of the model, there has been a significant improvement in accuracy, and the models exhibit strong generalization capabilities. Our research contributes to enhancing production efficiency, strengthening quality control, and improving material performance through computer vision technology. [ABSTRACT FROM AUTHOR]

Variable selection is one of the most important contents of machine learning research and a large number of variable selection methods have been proposed at present. From deep learning perspective, we propose a new variable selection method based on multi-layer artificial neural networks in this paper. The method propagates errors gained form a trained network model through the network from output layer to input layer and the values gained by input layer give information about the relative contribution. The variables with big contribution value are selected, which we call neural inverse propagation (NIP). Specifically, after training a neural network, the gained errors between the actual network outputs and the desired outputs go through the network from top to down and arrive at the input layer. The absolute values gained by input units provide information about the relative importance (contribution) of input units, which is that the larger the value is, the more important the corresponding input variable is. We remove k variables corresponding to the first k minimum absolute value gained by input units to the predictive variables, to perform variable selection. The efficiency of the proposed method is illustrated through its application to classification and regression tasks on a number of synthetic datasets and a real-world image classification dataset. [ABSTRACT FROM AUTHOR]

Hyperspectral remote sensing images (HSIs) have both spectral and spatial characteristics. The adept exploitation of these attributes is central to enhancing the classification accuracy of HSIs. In order to effectively utilize spatial and spectral features to classify HSIs, this paper proposes a method for the spatial feature extraction of HSIs based on a mutually guided image filter (muGIF) and combined with the band-distance-grouped principal component. Firstly, aiming at the problem that previously guided image filtering cannot effectively deal with the inconsistent information structure between the guided and target information, a method for extracting spatial features using muGIF is proposed. Then, aiming at the problem of the information loss caused by a single principal component as a guided image in the traditional GIF-based spatial–spectral classification, a spatial feature-extraction framework based on the band-distance-grouped principal component is proposed. The method groups the bands according to the band distance and extracts the principal components of each set of band subsets as the guide map of the current band subset to filter the HSIs. A deep convolutional neural network model and a generative adversarial network model for the filtered HSIs are constructed and then trained using samples for HSIs' spatial–spectral classification. Experiments show that compared with the traditional methods and several popular spatial–spectral HSI classification methods based on a filter, the proposed methods based on muGIF can effectively extract the spatial–spectral features and improve the classification accuracy of HSIs. [ABSTRACT FROM AUTHOR]

HyperGraph Neural Network (HGNN) has recently emerged as a promising approach for hyperspectral image classification (HSIC), reconciling state-of-the-art performance with powerful representation capabilities. However, existing HGNN-based methods have limited ability for hypergraph structure exploitation, leading to imperfect classification results. In this paper, we propose a framework called the residual enhanced hypergraph Neural Network (ResHGNN) to discover the potential structural features in hyperspectral image (HSI) data during deep neural networks. Specifically, ResHGNN first generates hyperedges from spatial-spectral features to construct a hypergraph representing fused spatial-spectral feature relationships in HSI. Then, the higher-order relationship among fused modal features is optimized by a residual enhanced hypergraph convolution learning process, to circumvent the HGNN-related over-smoothing issue. Experiments over three popular hyperspectral datasets show that the proposed classification method yields better performance than other models on the visual and numerical comparison. [ABSTRACT FROM AUTHOR]

The recognition of steel microstructure images plays a crucial role in the metallographic analysis process. Although some progress has been made through the application of artificial intelligence algorithms, several challenges remain. First, existing algorithms exhibit weak nonlinear feature extraction capabilities and noticeable limitations. Second, they overlook the intrinsic noise and redundant interference present in microscopic images. To address these issues, this paper investigates the automatic recognition of metallographic tissues by leveraging residual structures in deep neural networks. An enhanced residual network model based on transfer learning is proposed, which utilizes the pre-trained weights from the ImageNet dataset to facilitate learning with small sample data. This network offers higher classification accuracy and higher F1 scores. In addition, a deep residual shrinkage network model based on an attention mechanism is proposed. This model incorporates an attention sub-network into the original residual module and employs a soft threshold function to eliminate redundant features, including noise. The proposed algorithms are evaluated against various convolutional neural networks using 20 types of metallographic test sets. The experimental results showed that both methods have high accuracy rates of 95% and 94.44%, respectively, and F1 scores of 0.9464 and 0.9419. While maintaining the complexity of the model, there has been a significant improvement in accuracy, and the models exhibit strong generalization capabilities. Our research contributes to enhancing production efficiency, strengthening quality control, and improving material performance through computer vision technology. [ABSTRACT FROM AUTHOR]

This paper aims to propose an improved image classification model to reduce the cost of model construction. Aiming at the problem that network training usually requires the support of a large number of labeled samples, an image classification model based on semi-supervised deep learning is proposed, which uses labeled samples to guide the network to learn unlabeled samples. A convolutional neural network model for simultaneous processing of labeled and unlabeled data is constructed. The tagged data is used to train the Softmax classifier and provide the initial K-means clustering center for the untagged data. The nonsubsampling contourlet layer is used to replace the first convolutional layer of the full convolutional neural network to extract multi-scale depth features, and the nonsubsampling contourlet full convolutional neural network is constructed. The network can extract multi-scale information of the images to be classified, and extract more discriminative deep image features. In addition, the parameters of the nonsubsampled contourlet layers are pre-set and do not require network training. The proposed method has higher classification accuracy than the contrast method on polarimetric SAR images using the nonsubsampled contourlet full convolutional neural network. [ABSTRACT FROM AUTHOR]

This proposed research explores a novel approach to image classification by deploying a complex-valued neural network (CVNN) on a Field-Programmable Gate Array (FPGA), specifically for classifying 2D images transformed into polar form. The aim of this research is to address the limitations of existing neural network models in terms of energy and resource efficiency, by exploring the potential of FPGA-based hardware acceleration in conjunction with advanced neural network architectures like CVNNs. The methodological innovation of this research lies in the Cartesian to polar transformation of 2D images, effectively reducing the input data volume required for neural network processing. Subsequent efforts focused on constructing a CVNN model optimized for FPGA implementation, emphasizing the enhancement of computational efficiency and overall performance. The experimental findings provide empirical evidence supporting the efficacy of the image classification system developed in this study. One of the developed models, CVNN_128, achieves an accuracy of 88.3% with an inference time of just 1.6 ms and a power consumption of 4.66 mW for the classification of the MNIST test dataset, which consists of 10,000 frames. While there is a slight concession in accuracy compared to recent FPGA implementations that achieve 94.43%, our model significantly excels in classification speed and power efficiency—surpassing existing models by more than a factor of 100. In conclusion, this paper demonstrates the substantial advantages of the FPGA implementation of CVNNs for image classification tasks, particularly in scenarios where speed, resource, and power consumption are critical. [ABSTRACT FROM AUTHOR]

In recent years, deep neural networks have evolved rapidly in engineering technology, with models becoming larger and deeper. However, for most companies, developing large models is extremely costly and highly risky. Researchers usually focus on the performance of the model, neglecting its cost and accessibility. In fact, most regular business scenarios do not require high-level AI. A simple and inexpensive modeling method for fulfilling certain demands for practical applications of AI is needed. In this paper, a Fragmented neural network method is proposed. Inspired by the random forest algorithm, both the samples and features are randomly sampled on image data. Images are randomly split into smaller pieces. Weak neural networks are trained using these fragmented images, and many weak neural networks are then ensembled to build a strong neural network by voting. In this way, sufficient accuracy is achieved while reducing the complexity and data volume of each base learner, enabling mass production through parallel and distributed computing. By conducting experiments on the MNIST and CIFAR10 datasets, we build a model pool using FNN, CNN, DenseNet, and ResNet as the basic network structure. We find that the accuracy of the ensemble weak network is significantly higher than that of each base learner. Meanwhile, the accuracy of the ensemble network is highly dependent on the performance of each base learner. The accuracy of the ensemble network is comparable to or even exceeds that of the full model and has better robustness. Unlike other similar studies, we do not pursue SOTA models. Instead, we achieved results close to the full model with a smaller number of parameters and amount of data. [ABSTRACT FROM AUTHOR]

Deep neural networks have achieved remarkable results in the field of polarimetric synthetic aperture radar (PolSAR) image classification. However, PolSAR is affected by speckle imaging, resulting in PolSAR images usually containing a large amount of speckle noise, which usually leads to the poor spatial consistency of classification results and insufficient classification accuracy. Semantic segmentation methods based on deep learning can realize the task of image segmentation and classification at the same time, producing fine-grained and smooth classification maps. However, these approaches require enormous labeled data sets, which are laborious and time-consuming. Due to these issues, a new multi-modal contrastive fully convolutional network, named MCFCN, is proposed for PolSAR image classification in this paper, which combines multi-modal features of the same pixel as inputs to the model based on a fully convolutional network and accomplishes the classification task using only a small amount of labeled data through contrastive learning. In addition, to describe the PolSAR terrain targets more comprehensively and enhance the robustness of the classifier, a pixel overlapping classification strategy is proposed, which can not only improve the classification accuracy effectively but also enhance the stability of the method. The experiments demonstrate that compared with existing classification methods, the classification results of the proposed method for three real PolSAR datasets have higher classification accuracy. [ABSTRACT FROM AUTHOR]

We present a novel approach for automatically classifying illustrations from historical Chinese local gazetteers using modern deep learning techniques. Our goal is to facilitate the digital organization and study of a large quantity of digitized local gazetteers. We evaluate the performance of eight state-of-the-art deep neural networks on a dataset of 4,309 manually labeled and organized images of Chinese local gazetteer illustrations, grouped into three coarse categories and nine fine classes according to their contents. Our experiments show that DaViT achieved the highest classification accuracy of 93.9 per cent and F1-score of 90.6 per cent. Our results demonstrate the effectiveness of deep learning models in accurately recognizing and categorizing historical local gazetteer illustrations. We also developed a user-friendly web service to enable researchers easy access to the developed models. The potential for extending this method to other collections of scanned documents beyond Chinese local gazetteers makes a significant contribution to the study of visual materials in the arts and history in the digital humanities field. The dataset used in this study is publicly available and can be used for further research in the field. [ABSTRACT FROM AUTHOR]

Skin cancer is one of the most serious threats to human health among skin lesions. Computer-aided diagnosis methods can assist patients in identifying and detecting skin lesion types early, thereby enabling corresponding treatments. In this paper, we propose a dual-branch neural network model Conformer with Residual Cosine Similarity Attention and Bidirectional Convolutional fusion strategy, named RCSABC-Conformer. The core of this network structure comprises three parts: a Convolutional Neural Network (CNN) branch with Residual Cosine Similarity Attention (RCSA), a Transformer branch, and a Feature Couple Unit with Bidirectional Convolutional strategy (BC-FCU). The RCSA module calculates the cosine similarity value between the feature map generated by the convolutional operation and the feature map of the residual edge to assess whether their semantic information is similar. The semantic information of similar parts is weighted by exponential normalization to enhance the network's memory of similar features of the same type of skin lesion. The BC-FCU module interactively fuses local features and global representations of skin lesion images with different resolutions in the two branches. Specifically, when the global representations is integrated into local features, we introduce a new bidirectional convolution strategy to extract the feature map from both forward and backward directions, and then select the element with the smaller feature value from the two directions to fuse into local features. In this way, we can minimize the interference of the artifact features extracted by the Transformer branch on the CNN branch. In addition, taking advantage of the Transformer branch's capacity to construct global representations, our model can learn contextual semantic information of normal skin and lesion areas to enhance model robustness. We conducted experiments on three datasets, consisting of clinical and dermoscopic skin lesion images, as well as a hybrid of both. The experimental results show that RCSABC-Conformer outperforms both advanced and classical classification methods in terms of classification accuracy across all three datasets, without requiring an increase in the number of parameters and computational complexity. Compared with the baseline model, the classification accuracy of our proposed method improves by 2.40%, 5.39%, and 4.44% on the three datasets, respectively. To the best of our knowledge, this is the first study to apply an interactive fusion dual-branch network for multi-disease classification on different modalities of skin lesion databases. Code will be available at https://github.com/AlenLi817/RCSABC-Conformer. [ABSTRACT FROM AUTHOR]

Skin cancer is one of the most serious threats to human health among skin lesions. Computer-aided diagnosis methods can assist patients in identifying and detecting skin lesion types early, thereby enabling corresponding treatments. In this paper, we propose a dual-branch neural network model Conformer with Residual Cosine Similarity Attention and Bidirectional Convolutional fusion strategy, named RCSABC-Conformer. The core of this network structure comprises three parts: a Convolutional Neural Network (CNN) branch with Residual Cosine Similarity Attention (RCSA), a Transformer branch, and a Feature Couple Unit with Bidirectional Convolutional strategy (BC-FCU). The RCSA module calculates the cosine similarity value between the feature map generated by the convolutional operation and the feature map of the residual edge to assess whether their semantic information is similar. The semantic information of similar parts is weighted by exponential normalization to enhance the network's memory of similar features of the same type of skin lesion. The BC-FCU module interactively fuses local features and global representations of skin lesion images with different resolutions in the two branches. Specifically, when the global representations is integrated into local features, we introduce a new bidirectional convolution strategy to extract the feature map from both forward and backward directions, and then select the element with the smaller feature value from the two directions to fuse into local features. In this way, we can minimize the interference of the artifact features extracted by the Transformer branch on the CNN branch. In addition, taking advantage of the Transformer branch's capacity to construct global representations, our model can learn contextual semantic information of normal skin and lesion areas to enhance model robustness. We conducted experiments on three datasets, consisting of clinical and dermoscopic skin lesion images, as well as a hybrid of both. The experimental results show that RCSABC-Conformer outperforms both advanced and classical classification methods in terms of classification accuracy across all three datasets, without requiring an increase in the number of parameters and computational complexity. Compared with the baseline model, the classification accuracy of our proposed method improves by 2.40%, 5.39%, and 4.44% on the three datasets, respectively. To the best of our knowledge, this is the first study to apply an interactive fusion dual-branch network for multi-disease classification on different modalities of skin lesion databases. Code will be available at https://github.com/AlenLi817/RCSABC-Conformer. [ABSTRACT FROM AUTHOR]

Active learning alleviates labeling costs by selecting and labeling the most informative examples from an unlabeled pool. However, most existing active learning approaches estimate informativeness with uncalibrated confidence, resulting in unreliable informativeness estimation. These approaches generally ignored two significant issues caused by uncalibrated confidence methods. Firstly, the average uncalibrated confidence generated by modern neural networks is usually higher than the accuracy. Secondly, examples located near the decision boundaries are unstable during prediction when the target model updates parameters in the last several epochs, even throughout the training process. This phenomenon, caused by the forgetting characteristic of neural networks, has a significant impact on some specific models that estimate the informativeness by predicted probability vectors or pseudo labels. To address these issues, in this paper, we propose a novel active learning approach to reliably estimate informativeness with calibrated confidence. Specifically, we integrate the intermediate predictions for each unlabeled example, generated by the target model during the training process, to generate calibrated confidence. The calibrated confidence can capture a tendentious label from an indecisive subset of the class space. We show that the calibrated confidence with tendentiousness can maintain the ability of correct predictions. The empirical results demonstrate that our approach outperforms the state-of-the-art active learning methods on image classification tasks. • The active learning algorithm selects unstable examples by calibrated confidence.. • The confidence calibration method only requires training a single model. • Extensive experiments show the competitive performance of the proposed approach. [ABSTRACT FROM AUTHOR]

Adversarial training has become the mainstream method to boost adversarial robustness of deep models. However, it often suffers from the trade-off dilemma, where the use of adversarial examples hurts the standard generalization of models on natural data. To study this phenomenon, we investigate it from the perspective of spatial attention. In brief, standard training typically encourages a model to conduct a comprehensive check to input space. But adversarial training often causes a model to overly concentrate on sparse spatial regions. This reduced tendency is beneficial to avoid adversarial accumulation but easily makes the model ignore abundant discriminative information, thereby resulting in weak generalization. To address this issue, this paper introduces an Attention-Enhanced Learning Framework (AELF) for robustness training. The main idea is to enable the model to inherit the attention pattern of standard pre-trained model through an embedding-level regularization. To be specific, given a teacher model built on natural examples, the embedding distribution of teacher model is used as a static constraint to regulate the embedding outputs of the objective model. This design is mainly supported with that the embedding feature of standard model is usually recognized as a rich semantic integration of input. For implementation, we present a simplified AELFs that can achieve the regularization with single cross entropy loss via the parameter initialization and parameter update strategy. This avoids the extra consistency comparison operation between embedding vectors. Experimental observations verify the rationality of our argument, and experimental results demonstrate that it can achieve remarkable improvements in generalization under the high-level robustness. [ABSTRACT FROM AUTHOR]

Deep neural networks tend to suffer from the overfitting issue when the training data are not enough. In this paper, we introduce two metrics from the intra-class distribution of correct-predicted and incorrect-predicted samples to provide a new perspective on the overfitting issue. Based on it, we propose a knowledge distillation approach without pretraining a teacher model in advance named Tolerant Self-Distillation (TSD) for alleviating the overfitting issue. It introduces an online updating memory and selectively stores the class predictions of the samples from the past iterations, making it possible to distill knowledge across the iterations. Specifically, the class predictions stored in the memory bank serve as the soft labels for supervising the samples from the same class for the current iteration in a reverse way, i.e. the correct-predicted samples are supervised with the incorrect predictions while the incorrect-predicted samples are supervised with the correct predictions. Consequently, the premature convergence issue caused by the over-confident samples would be mitigated, which helps the model to converge to a better local optimum. Extensive experimental results on several image classification benchmarks, including small-scale, large-scale, and fine-grained datasets, demonstrate the superiority of the proposed TSD. [ABSTRACT FROM AUTHOR]

Gait recognition is an important biometric technology that allows for the remote collection of stakeholders' characteristics, without requiring their explicit cooperation. It has gained considerable attention in the fields of criminal investigation and intelligent security. Previous studies have shown that local gait features can enhance gait recognition performance by improving robustness to disturbances. However, global gait features also play a crucial role in gait recognition. Many researchers have utilized convolutional operations to extract global features, but these operations tend to focus on features within the receptive field, neglecting those outside of it. Therefore, the potential of global gait features has not been fully explored. In this paper, we propose a gait recognition framework based on vision transformers, aiming to enhance the extraction of global gait features. We introduce an adaptive multi-frame global feature mapping (AMGM) method to address the challenge of inconsistent feature dimensions caused by variations in the number of gait frames when fusing global and local features. We evaluate our model on the latest datasets, and the experimental results demonstrate a significant breakthrough. Notably, our model achieves state-of-the-art recognition accuracy, particularly in scenarios where subjects are wearing coats. Additionally, our model achieves remarkable improvements in recognition accuracy through training with small sample sets. [ABSTRACT FROM AUTHOR]

An increasing need of running Convolutional Neural Network (CNN) models on mobile devices encourages the studies on efficient and lightweight neural network model. In this paper, an Inverse Residual Multi-Branch Network named IremulbNet is proposed to solve the problem of insufficient classification accuracy in existing lightweight network models. The core module of this model is to reconstruct an inverse residual structure, in which a special feature fusion method, multi-branch feature extraction, and depthwise separable convolution techniques are used to improve the classification accuracy. The performance of model is tested using image databases. Experimental results show that for the fine-grained image dataset Imagenet-woof, IremulbNet achieved 10.9%, 12.2%, and 15.3% higher accuracy than that of MobileNet V3, ShuffleNet V2, and PeleeNet, respectively. Moreover, it can reduce inference time (GPU) about 42.09% and 75.56% compared to classic ResNet50 and DenseNet121. • A lightweight IremulbNet is proposed to achieve high accuracy image recognition with a small number of parameters and operations. • A special feature fusion method, multi-branch feature extraction, and depthwise separable convolution are used to improve the performance of the network. • A significant improvement in recognition accuracy can be obtained in image classification tasks with low computation load. [ABSTRACT FROM AUTHOR]

• As the steep switching device has double-gate, the threshold voltage can be adaptively controlled. • The adjustment of the threshold can be employed to mitigate the accuracy deterioration. • Off current of proposed integrate and fire neuron circuit decreases significantly. • Adjustable threshold of the neuron circuit can achieve a high inference accuracy. Deep learning has shown impressive capabilities in tasks like speech recognition and image classification. However, modern deep neural networks often demand a significant number of weights and extensive computational resources, creating efficiency challenges for applications on edge devices. To address these issues, researchers have introduced deep spiking neural networks (DSNNs) that leverage specialized hardware for synapses and neurons. DSNNs offer a potential solution by improving efficiency in edge-device applications. In this paper, the hardware based DSNN with integrate and fire neuron using steep switching device was investigated. We propose integrate and fire neuron using steep switching device to implement rate coding as input encoding method. Because the steep switching device has double-gate, the threshold voltage of the neuron circuits can be adaptively controlled, which changes the rates of input pulse. Hence, the adjustment of the threshold of neuron can be employed to mitigate the accuracy deterioration resulting from the transformation from deep neural networks (DNNs) to DSNNs. In addition, the off-current of proposed integrate and fire neuron circuit decreases significantly as the steep switching device has steep subthreshold swing. A system simulation of a hardware based DSNN shows that the adjustable threshold of the neuron circuit can achieve a high inference accuracy of 98.36 % which is comparable to that obtained with software based DNN. [ABSTRACT FROM AUTHOR]

In this paper, we present a novel dimensionality reduction term (i.e. TripletPCA and ContrastivePCA) designed to enhance the robustness of pair-based loss functions in metric learning against image corruptions. Notably, our approach achieves this without the need for any data preprocessing steps. Our method can be seamlessly integrated into existing models, such as a block with Plug & Play framework. To the best of our knowledge, our TripletPCA and ContrastivePCA represent the first attempts to incorporate dimensionality reduction directly into pair-based metric learning losses for end-to-end metric learning. By projecting image features into low-dimensional vectors, our proposed loss functions effectively retain the essential components of images while mitigating the impact of corrupted features. Consequently, our metric learning loss function accurately computes feature distances through the projection. Experimental results demonstrate that our dimensionality reduction term can be easily incorporated into various types of existing deep neural networks. This integration leads to a substantial improvement in performance on standard benchmark datasets for corrupted image classification tasks. On several corruption datasets, we achieve an average performance improvement of 10.55% compared to existing baseline methods. Our code is available at https://github.com/Juryun/TDRL. [ABSTRACT FROM AUTHOR]

Spiking neural networks (SNNs) have garnered significant attention due to their ultra-high-speed and ultra-low-power operation, rendering them suitable for a range of energy-efficient applications. However, their performance in image generation is relatively mediocre, resulting in a lack of an effective generative replay mechanism to handle sequential probabilistic representations of multiple datasets. Consequently, this limitation often leads to catastrophic forgetting. This paper introduces a novel SNN lifelong learning framework namely Dynamic Lifelong learning with Spiking Generative Networks (DL-SGN), aimed at providing better generation ability while mitigating the problem of catastrophic forgetting in a continual learning scenario. DL-SGN comprises three key components: dynamic experts, a student, and an assistant. The dynamic expert is implemented as a dynamically expanding mixture model, with a proposed network expansion mechanism Dynamic Knowledge Adversarial Fusion (DKAF) facilitating the automatic handling of an increasing number of tasks. the student module adopts an SNN-based Variational Autoencoder (VAE) and classifier, leveraging accumulated knowledge from each expert. To enhance the student's ability to generalize to images, we introduce a discriminator as an assistant module trained using adversarial training. We validate the effectiveness of the proposed by conducting experiments on image generation and classification in a lifelong learning environment, and the results substantiate the effectiveness of DL-SGN. • The DL-SGN framework is proposed to addresses catastrophic forgetting in SNNs. • Our model is based on multiple dynamic experts and adversarial training. • Develop a dynamic knowledge adversarial fusion to expand expert module's capacity. • Employ an assistant module to enhance the student module's generalization ability. [ABSTRACT FROM AUTHOR]

Model pruning is a commonly used technique for compressing DNNs and reducing computation requirements to accelerate inference. However, the required granularity of pruning varies across different application scenarios, making it difficult and cumbersome to customize different pruning methods for each hardware or computing platform. Therefore, a unified framework is necessary to accommodate various levels of granularity of pruning. Furthermore, some available methods require additional fine-tuning or model retraining to restore accuracy, which can result in significant time costs. With this motivation, this paper proposes a Multi-Granularity Pruning Framework, namely MGPF, to obtain sparse models of different granularity without fine-tuning the remaining connections. Specifically, a deformable soft mask is introduced as the pruning initiator to achieve different levels of pruning granularity, such as weight pruning, channel pruning, and filter pruning, etc. The model parameters and soft masks are jointly trained, and we just apply L 1 regularization on soft masks for sparsity to ensure that the model can be repaired during training without fine-tuning or retraining. After pruning, the soft masks are absorbed into the model parameters in the form of element product without changing the model structures. Experimental results on three image classification benchmarks CIFAR-10/100 and ImageNet-1K demonstrate the effectiveness of our method for various CNN architectures, datasets, and pruning rates. Particularly, for ResNet-50 on ImageNet-1K, we achieve a higher accuracy under the pruning rate of 98% for unstructured pruning which leads the advanced method by 12%. [ABSTRACT FROM AUTHOR]

Deep Neural Networks have received considerable attention in recent years. As the complexity of network architecture increases in relation to the task complexity, it becomes harder to manually craft an optimal neural network architecture and train it to convergence. As such, Neural Architecture Search (NAS) is becoming far more prevalent within computer vision research, especially when the construction of efficient, smaller network architectures is becoming an increasingly important area of research, for which NAS is well suited. However, despite their promise, contemporary and end-to-end NAS pipeline require vast computational training resources. In this paper, we present a comprehensive overview of contemporary NAS approaches with respect to image classification, object detection, and image segmentation. We adopt consistent terminology to overcome contradictions common within existing NAS literature. Furthermore, we identify and compare current performance limitations in addition to highlighting directions for future NAS research. [ABSTRACT FROM AUTHOR]

Recently, convolutional neural networks (CNNs), which obtain powerful deep features in an end-to-end manner, have achieved powerful performance in remote sensing scene classification. However, the average or maximum pooling operations defined in the spatial domain and coarser-resolution features with high levels cannot extract reliable features and clear boundaries for small-scale targets in remote sensing scene imagery. This paper attempts to address these problems and proposes a multi-domain sematic high-order network for scene classification, named MSHNet. First, wavelet-spatial and detachable pooling blocks defined in the wavelet and spatial domains are inserted at the end of the convolutional block to learn the features in a more structural fusion manner. Second, multi-scale and multi-resolution semantic embedding modules are proposed to take full advantage of the complementary information and effectively maintain the spatial structures of learned deep features. Third, we employ a factorized bilinear coding approach to obtain compact and discriminative second-order features. MSHNet is thoroughly evaluated on two publicly available benchmarks, i.e., AID (Aerial Image Dataset) and NWPU-RESISC45 (Northwestern Polytechnical University-Remote Sensing Image Scene Classification 45). The extensive results illustrate that our MSHNet is competitive with other related multi-scale deep neural networks. [Display omitted] • Wavelet-spatial and detachable pooling maintain structures of learned features. • Factorized bilinear coding method with semantic features captures rich statistics. • MSHNet obtains competitive performance on NWPU-RESISC45 and AID datasets. [ABSTRACT FROM AUTHOR]

Traditional knowledge distillation approaches are typically designed for specific tasks, as they primarily distilling deep features from intermediate layers of a neural network, generally with ingeniously designed knowledge representations, which increase the difficulty of model development and interpretation. In contrast, we empirically show that a soft logits knowledge distillation is enough to significantly narrow the teacher–student model performance gap. In this paper, the S oft H ybrid K nowledge D istillation (SHKD) is proposed to generate representative features from the input and output location of models, and applicable to various teacher and student architectures. Specifically, the additional mixup task is first applied to the same batch of training images, which enriches the high-order feature information transferred by the model. And then, a top-k guided specification filter is added to the teacher model, to reduce the bias induced throughout the knowledge transfer process. Finally, the SHKD utilizes the same and different teacher–student model pairs of the deep neural network, which can be easily adjusted to the computational resources available in the current world. Extensive experiments are conducted to examine the efficiency, and show that the SHKD outperforms state-of-the-art knowledge distillation methods. With the SHKD, the high performance of models will be maintained, and the gap between the teacher model and the student model can be reduced, making the construction of deep neural networks more convenient to the unknown scenarios. The source code is available at https://github.com/lambett/SHKD. [ABSTRACT FROM AUTHOR]

Recent improvements in dynamic hand gesture recognition have seen a shift from traditional convolutional architectures to attention-based networks. These attention networks have been proven to outclass CNN + LSTM architectures, showing higher accuracy as well as reduced model parameters. Especially, skeleton-based attention networks have been shown to outperform visual-based networks due to the rich information from skeleton-based hand features. However, there is an opportunity to introduce complementary features from other modalities like RGB, depth, and optical flow images to enhance the recognition capability of skeleton-based networks. This paper aims to explore the addition of a multimodal fusion network to a skeleton-based Hierarchical Self-Attention Network (MF-HAN) and test for increased model effectiveness. Unlike traditional fusion techniques, this fusion network uses features derived from other sources of multimodal data in a reduced feature space using a cross-attention layer. The model outperforms its root model and other state-of-the-art models on the SHREC'17 track dataset, especially in the 28 gestures setting by more than 1 % in gesture classification accuracy. The experimentation was tested on the DHG dataset as well. [ABSTRACT FROM AUTHOR]

Activation functions play an important role in deep learning models by introducing non-linearity to the output of a neuron, enabling the network to learn complex patterns and non-linear relationships in data and make predictions on more complex tasks. Deep learning models' most commonly used activation functions are Purelin, Sigmoid, Tansig, Rectified Linear Unit (ReLU), and Exponential Linear Unit (ELU), which exhibit limitations such as non-differentiability, vanishing gradients, and neuron inactivity with negative values. These functions are typically defined over a finite range, and their outputs are integers or real numbers. Using fractional calculus in designing activation functions for neural networks has shown promise in improving the performance of deep learning models in specific applications. These activation functions can capture more complex non-linearities than traditional integer-order activation functions, improving performance on tasks such as image classification and time series prediction. This paper focuses on deriving and testing linear and non-linear fractional-order forms of activation functions and their variants. The linear activation function includes Purelin. In contrast, the non-linear activation functions are Binary Step, Sigmoid, Tansig, ReLU, ELU, Gaussian Error Linear Unit (GELU), Hexpo, and their variants. Besides, the standard formula has been implemented and used in developing the fractional-order linear activation function. Furthermore, various expansion series, such as Euler and Maclaurin, have been used to design non-linear fractional-order activation functions and their variants. The single- and multi-layer fractional-order neural network models have been developed using the designed fractional-order activation functions. The simulation study uses developed fractional-order neural network models for predicting the Texas wind turbine systems' generated power. The performance of single and multi-layer fractional-order neural network models has been evaluated by changing the activation functions in the hidden layer while keeping the Purelin function constant at the output layer. Experiments on neural network models demonstrate that the designed fractional-order activation functions outperform traditional functions like Sigmoid, Tansig, ReLU, ELU, and their variants, effectively addressing limitations. • Activation functions are derived into fractional functions using fractional calculus. • The fractional neural network models have been made by adopting derived functions. • A Case study has been done on the Texas wind turbine with the models to predict the power. • The models' performance has been evaluated using the fractional functions. [ABSTRACT FROM AUTHOR]

This paper discusses how automatic speech recognition systems are and could be designed, in order to best exploit the discriminative information encoded in human speech. This contrasts with many recent machine learning approaches that apply general recognition architectures to signals to identify, with little concern for the nature of the input. The implicit assumption has often been that training can automatically discover the useful properties that exist in signals, with minimal manual intervention. These approaches may be suitable for some tasks such as image recognition, where the diversity of visual input is vast; e.g., an image may be any (natural or synthetic) scene that a camera views. We first examine what makes speech special, i.e., a natural signal from a complex tube, driven by a source that is quasi-periodic and/or noisy, aiming to communicate a wide variety of information, using the different vocal systems of human speakers. Then, we view how pertinent features are extracted from speech via efficient means, related to the objectives of communication. We see how to reliably and efficiently identify the different units of oral language. We learn from the history of attempts to do ASR, e.g., why they succeeded and how improved methods exploited the increasing availability of data and computer power (in particular, deep neural networks). Finally, we suggest ways to render ASR both more accurate and efficient. This work is aimed at both newcomers to ASR and experts, in terms of presenting issues broadly, but without mathematical or algorithmic details, which are readily found in the references. [ABSTRACT FROM AUTHOR]

Domain adaptation provides the possibility of utilizing the knowledge gained from an auxiliary domain to accomplish the task in another related domain. In this paper, a common feature representation of these domains is learned through a new proposed idea to reduce inter-domain differences more precisely which results in higher accuracy for unsupervised domain adaptation. To decrease these divergences, the proposed method finds a subspace that reduces data distribution discrepancy between domains. To achieve this goal, minimization of the divergence of the class conditional probability distributions is considered. By considering class conditional distributions, more discriminative information of data is preserved compared to applying marginal distributions. Also, an explicit parametric distribution of the source and target domains is considered to reduce the discrepancy between the two domains data which results in higher accuracy compared to the other relevant domain adaptation methods. Experimental studies on benchmark image classification tasks confirmed our assumptions and indicated the significant improvement given by the proposed method compared to the other state-of-the-art methods. • Aligning the feature distribution of the source and target domains explicitly. • Introducing a new loss function to impose the common distribution on the features. • Reducing the class conditional probability distribution distance between the domains. [ABSTRACT FROM AUTHOR]