Search

Machine learning is increasingly used as a computing paradigm in cartographic research. In this extended editorial, we provide some background of the papers in the CaGIS special issue Machine Learning in Cartography with a special focus on pattern recognition in maps, cartographic generalization, style transfer, and map labeling. In addition, the paper includes a discussion about map encodings for machine learning applications and the possible need for explicit cartographic knowledge and procedural modeling in cartographic machine learning models. [ABSTRACT FROM AUTHOR]

Objectives: Recognition of building group patterns is an important part of the building generalization. Efficient methods for the recognition of building group patterns can effectively improve the quality of automated map generalization. Traditional recognition methods mainly include geometric methods and traditional machine learning methods, which are limited by the complex rule definitions and huge feature engineering. The graph convolution neural network (GCN) can overcome the limitations of traditional methods to some extent and has been successfully applied to the pattern analysis of buildings. However, there are few methods to recognize multiple building group patterns using GCN, and the existing GCN methods are almost based on spectral graph convolution, which do not consider local spatial information fully. Methods: As the spatial GCN is more efficient than the spectral GCN, this paper introduces a spatial GCN method DGCNN to recognize three building group patterns, including linear pattern, grid pattern and irregular pattern. The first step is to cluster the buildings into different groups, then some indices are selected to construct the feature vector of each group according to the Gestalt principles. Second, Delaunay triangulation (DT) and minimum spanning tree (MST) are chosen to construct graph structures of building groups. Finally, graph structures are used as input for the DGCNN model, and building group patterns are obtained after training. Results: The experiment selects the urban areas in Shanghai and we compare the accuracies with traditional spectral GCN method. The results show that the accuracy for the DGCNN model in the train set can reach 97.06%, and 95% for the test set. And the accuracies for the spectral GCN model in the train set is 90.90%, and 86.05% in the test set. Conclusions: The proposed method overcomes the limitations of spectral GCN, and it does not need graph Fourier transform. It can improve the recognition accuracy significantly and faces less difficulty in explaining the process of feature extraction. Hence it is an effective method for the recognition of building group patterns. [ABSTRACT FROM AUTHOR]

In the application areas of streaming, social networks, and video-sharing platforms such as YouTube and Facebook, along with traditional television systems, programs’ classification stands as a pivotal effort in multimedia content management. Despite recent advancements, it remains a scientific challenge for researchers. This paper proposes a novel approach for television monitoring systems and the classification of extended video content. In particular, it presents two distinct techniques for program classification. The first one leverages a framework integrating Structural Similarity Index Measurement and Convolutional Neural Network, which pipelines on stacked frames to classify program initiation, conclusion, and contents. Noteworthy, this versatile method can be seamlessly adapted across various systems. The second analyzed framework implies directly processing optical flow. Building upon a shot-boundary detection technique, it incorporates background subtraction to adaptively discern frame alterations. These alterations are subsequently categorized through the integration of a Transformers network, showcasing a potential advancement in program classification methodology. A comprehensive overview of the promising experimental results yielded by the two techniques is reported. The first technique achieved an accuracy of 95%, while the second one surpassed it with an even higher accuracy of 87% on multiclass classification. These results underscore the effectiveness and reliability of the proposed frameworks, and pave the way for a more efficient and precise content management in the ever-evolving landscape of multimedia platforms and streaming services. [ABSTRACT FROM AUTHOR]

Speech emotion recognition is of great significance in the industry such as social robots, health care, and intelligent education. Due to the obscurity of emotional expression in speech, most works on speech emotion recognition (SER) ignore the consistency of speech emotion recognition, leading to fuzzy expression and low accuracy in emotional recognition. In this paper, we propose a semantic aware speech emotion recognition model to alleviate this issue. Specifically, a speech feature extraction module based on CNN and Transformer is designed to extract local and global information from the speech. Moreover, a semantic embedding support module is proposed to use text semantic information as auxiliary information to assist the model in extracting emotional features of speech, and can effectively overcome the problem of low recognition rate caused by emotional ambiguity. In addition, the model uses a key-value pair attention mechanism to fuse the features, which makes the fusion of speech and text features preferable. In experiments on two benchmark corpora IEMOCAP and EMO-DB, the recognition rates of 74.3% and 72.5% were obtained under respectively, which show that the proposed model can significantly improve the accuracy of emotion recognition. [ABSTRACT FROM AUTHOR]

License plate detection is an important task in Intelligent Transportation Systems (ITS) and has a wide range of applications in vehicle management, traffic control, and public safety. In order to improve the accuracy and speed of mobile recognition, an improved lightweight YOLOv5s model is proposed for license plate detection. First, an improved Stemblock network is used to replace the original Focus layer in the network, which ensures strong feature expression capability and reduces a large number of parameters to lower the computational complexity; then, an improved lightweight network, ShuffleNetv2, is used to replace the backbone network of the YOLOv5s, which makes the model lighter and ensures the detection accuracy at the same time. Then, a feature enhancement module is designed to reduce the information loss caused by the rearrangement of the backbone network channels, which facilitates the information interaction in the feature fusion process; finally, the low‐, medium‐ and high‐level features in the Shufflenetv2 network structure are fused to form the final high‐level output features. Experimental results on the CCPD dataset show that compared to other methods this paper obtains better performance and faster speed in the license plate detection task, in which the average precision mean value reaches 96.6%, and can achieve a detection speed of 43.86 frame/s, and the parameter volume is reduced to 5.07 M. [ABSTRACT FROM AUTHOR]

Pattern recognition of linear feature sets, such as river networks, road networks, and contour clusters, is essential in cartography and geographic information science. Previous studies have investigated many methods to identify the patterns of linear feature sets; the key to each of these studies is to generate a reasonable and computable representation for each set. However, most existing methods are only designed for a specific task or data type and cannot provide a general solution for formalizing linear feature sets owing to their complex geometric characteristics, spatial relations and distributions. In addition, some methods require human involvement to specify characteristics, choose parameters, and determine the weights of different measures. To reduce human intervention and improve adaptability to various feature types, this paper proposes a novel deep learning architecture for learning the representations of linear feature sets. The presented model accepts vector data directly without extra data conversion and feature extraction. After generating local, neighborhood, and global representations of inputs, the representations are aggregated accordingly to perform pattern recognition tasks, including classification and segmentation. In the experiments, building groups classification and road interchanges segmentation achieved accuracies of 98% and 89%, respectively, indicating the model's effectiveness and adaptability. [ABSTRACT FROM AUTHOR]

• Importance of gait recognition for security and surveillance applications. • Comparison of ML and DL methods in terms of effectiveness. • Examples of specific applications, such as fall detection and biometric identification. • Challenges and limitations of current methods. • Future directions and potential improvements. The study of gait recognition, a biometric application that identifies individuals based on their unique walking patterns, is an evolving field. In this paper, we conduct a literature review to compare the performance of machine learning and deep learning approaches in covariate conditions, focusing on the specific aspects of deep learning pipelines in gait recognition. We highlight commonly used strategies and open problems in identification based on behavioral traits and propose future perspectives for researchers in this field. Through our investigation, we aim to provide insights that will aid researchers in developing informed decisions when to take which data preprocessing technique in designing gait recognition systems. Our paper provides a comprehensive exposition of machine learning versus deep learning architectures and pipelines for biometric applications using human gait and serves as a valuable resource for researchers in this area. [ABSTRACT FROM AUTHOR]

Gait recognition, also known as walking pattern recognition, has expressed deep interest in the computer vision and biometrics community due to its potential to identify individuals from a distance. It has attracted increasing attention due to its potential applications and non-invasive nature. Since 2014, deep learning approaches have shown promising results in gait recognition by automatically extracting features. However, recognizing gait accurately is challenging due to the covariate factors, complexity and variability of environments, and human body representations. This paper provides a comprehensive overview of the advancements made in this field along with the challenges and limitations associated with deep learning methods. For that, it initially examines the various gait datasets used in the literature review and analyzes the performance of state-of-the-art techniques. After that, a taxonomy of deep learning methods is presented to characterize and organize the research landscape in this field. Furthermore, the taxonomy highlights the basic limitations of deep learning methods in the context of gait recognition. The paper is concluded by focusing on the present challenges and suggesting several research directions to improve the performance of gait recognition in the future. [ABSTRACT FROM AUTHOR]

Some fundamental visual features have been found to be fully extracted before reaching the cerebral cortex. We focus on direction-selective ganglion cells (DSGCs), which exist at the terminal end of the retinal pathway, at the forefront of the visual system. By utilizing a layered pathway composed of various relevant cells in the early stage of the retina, DSGCs can extract multiple motion directions occurring in the visual field. However, despite a considerable amount of comprehensive research (from cells to structures), a definitive conclusion explaining the specific details of the underlying mechanisms has not been reached. In this paper, leveraging some important conclusions from neuroscience research, we propose a complete quantified model for the retinal motion direction selection pathway and elucidate the global motion direction information acquisition mechanism from DSGCs to the cortex using a simple spiking neural mechanism. This mechanism is referred to as the artificial visual system (AVS). We conduct extensive testing, including one million sets of two-dimensional eight-directional binary object motion instances with 10 different object sizes and random object shapes. We also evaluate AVS's noise resistance and generalization performance by introducing random static and dynamic noises. Furthermore, to thoroughly validate AVS's efficiency, we compare its performance with two state-of-the-art deep learning algorithms (LeNet-5 and EfficientNetB0) in all tests. The experimental results demonstrate that due to its highly biomimetic design and characteristics, AVS exhibits outstanding performance in motion direction detection. Additionally, AVS possesses biomimetic computing advantages in terms of hardware implementation, learning difficulty, and parameter quantity. [ABSTRACT FROM AUTHOR]

The recognition of human activities using vision-based techniques has become a crucial research field in video analytics. Over the last decade, there have been numerous advancements in deep learning algorithms aimed at accurately detecting complex human actions in video streams. While these algorithms have demonstrated impressive performance in activity recognition, they often exhibit a bias towards either model performance or computational efficiency. This biased trade-off between robustness and efficiency poses challenges when addressing complex human activity recognition problems. To address this issue, this paper presents a computationally efficient yet robust approach, exploiting saliency-aware spatial and temporal features for human action recognition in videos. To achieve effective representation of human actions, we propose an efficient approach called the dual-attentional Residual 3D Convolutional Neural Network (DA-R3DCNN). Our proposed method utilizes a unified channel-spatial attention mechanism, allowing it to efficiently extract significant human-centric features from video frames. By combining dual channel-spatial attention layers with residual 3D convolution layers, the network becomes more discerning in capturing spatial receptive fields containing objects within the feature maps. To assess the effectiveness and robustness of our proposed method, we have conducted extensive experiments on four well-established benchmark datasets for human action recognition. The quantitative results obtained validate the efficiency of our method, showcasing significant improvements in accuracy of up to 11% as compared to state-of-the-art human action recognition methods. Additionally, our evaluation of inference time reveals that the proposed method achieves up to a 74× improvement in frames per second (FPS) compared to existing approaches, thus showing the suitability and effectiveness of the proposed DA-R3DCNN for real-time human activity recognition. [ABSTRACT FROM AUTHOR]

With the vigorous development of the Internet, multimedia technology, and virtual reality technology, digital museum has been born and gradually entered the stage of rapid development. As a typical application of 3D virtual scene in the field of traditional culture protection and communication, 3D virtual museum has become an important trend in the development of museums in recent years. Three-dimensional virtual museum can maximize the development of museum functions and better play the museum's cultural communication and heritage protection and other functions. Faced with such rich cultural relic resources, how to construct 3D virtual museum quickly becomes an important research problem. Virtual reality technology integrates the latest research and development achievements of digital image graphics processing technology, computer technology, artificial intelligence, multimedia technology, sensor technology, and other information technologies, which provides powerful support and help for us to create and experience virtual world. The combination of virtual reality technology and cultural heritage protection has created a new research field. Based on this, this paper studies personalized healthcare museum exhibition system design based on the VR and deep learning-driven multimedia and multimodal sensing. The novel computational model is designed to improve the systematic performance of the VR framework, and the application scenarios are tested. Through the verification, the performance is tested. [ABSTRACT FROM AUTHOR]

Camouflaged object detection (COD), in a broad sense, aims to detect image objects that have high degrees of similarity to the background. COD is more challenging than conventional object detection because of the high degree of "fusion" between a camouflaged object and the background. In this paper, we focused on the accurate detection of camouflaged objects, conducting an in-depth study on COD and addressing the common detection problems of high miss rates and low confidence levels. We proposed a ternary cascade perception-based method for detecting camouflaged objects and constructed a cascade perception network (CPNet). The innovation lies in the proposed ternary cascade perception module (TCPM), which focuses on extracting the relationship information between features and the spatial information of the camouflaged target and the location information of key points. In addition, a cascade aggregation pyramid (CAP) and a joint loss function have been proposed to recognize camouflaged objects accurately. We conducted comprehensive experiments on the COD10K dataset and compared our proposed approach with other seventeen-object detection models. The experimental results showed that CPNet achieves optimal results in terms of six evaluation metrics, including an average precision (AP)50 that reaches 91.41, an AP75 that improves to 73.04, and significantly higher detection accuracy and confidence. [ABSTRACT FROM AUTHOR]

Pattern recognition based on a high-dimensional predictor is considered. A classifier is defined which is based on a Transformer encoder. The rate of convergence of the misclassification probability of the classifier towards the optimal misclassification probability is analyzed. It is shown that this classifier is able to circumvent the curse of dimensionality provided the a posteriori probability satisfies a suitable hierarchical composition model. Furthermore, the difference between the Transformer classifiers theoretically analyzed in this paper and the ones used in practice today is illustrated by means of classification problems in natural language processing. [ABSTRACT FROM AUTHOR]

In recent years, with the diversification of people's modes of transportation, a large amount of traffic data is generated when people travel every day, and this data can help transportation mode detection to be of great use in a variety of applications. Although transportation mode detection has been investigated, there are still challenges in terms of accuracy and robustness. This paper presents a novel transportation mode detection algorithm, DFTrans, which is based on Temporal Block and Attention Block. Low- and high-frequency components of traffic sequences are obtained using discrete wavelet transforms. A two-channel encoder is carefully designed to accurately capture the temporal and spatial correlation between low- and high-frequency components in both long- and short-term patterns. With the Temporal Block, the inductive bias of the CNN is introduced at high frequencies to improve generalization performance. At the same time, the network is generated with the same length as the input, ensuring a long effective history. Low frequencies are passed through Attention Block, which has fewer parameters to capture the global focus and solves the problem that RNNs cannot be computed in parallel. After fusing the output of the feature by Temporal Block and Attention Block, the classification results are output by MLP. Extensive experimental results show that the DFTrans algorithm achieves macro F1 scores of 86.34% on the real-world SHL dataset and 87.64% on the HTC dataset. Our model can better identify eight modes of transportation, including stationary, walking, running, cycling, bus, car, underground, and train, and has better performance in transportation mode detection than other baseline algorithms. [ABSTRACT FROM AUTHOR]

Credit card transactions may contain some categorical attributes with large domains, involving up to hundreds of possible values, also known as high-cardinality attributes. The inclusion of such attributes makes analysis harder, due to results with poorer generalization and higher resource usage. A common practice is, therefore, to ignore such attributes, removing them, albeit wasting the information they provided. Contrariwise, this paper reports our findings on the positive impacts of using high-cardinality attributes on credit card fraud detection. Thus, we present a new algorithm for domain reduction that preserves the fraud-detection capabilities. Experiments applying a deep feedforward neural network on real datasets from a major Brazilian financial institution have shown that, when measured by the F-1 metric, the inclusion of such attributes does improve fraud-detection quality. As a main contribution, this proposed algorithm was able to reduce attribute cardinality, improving the training times of a model while preserving its predictive capabilities. [ABSTRACT FROM AUTHOR]

Nowadays, political parties have widely adopted social media for their party promotions and election campaigns. During the election, Twitter and other social media platforms are used for political coverage to promote the party and its candidates. This research discusses and estimates the stability of many volumetric social media approaches to forecast election results from social media activities. Numerous machine learning approaches are applied to opinions shared on social media for predicting election results. This paper presents a machine learning model based on sentiment analysis to predict Pakistan's general election results. In a general election, voters vote for their favorite party or candidate based on their personal interests. Social media has been extensively used for the campaign in Pakistan general election 2018. Using a machine learning technique, we provide a five-step process to analyze the overall election results, whether fair or unfair. The work is concluded with detailed experimental results along with discussion on the outcomes of sentiment analysis for real-world forecasting and approval of general elections in Pakistan. [ABSTRACT FROM AUTHOR]

The unceasing evolution of analytical instrumentation determines an exponential increase of data production, which in turn boosts new cutting-edge analytical challenges, requiring a progressive integration of artificial intelligence (AI) algorithms into the instrumental data treatment software. Machine learning, deep learning, and computer vision are the most common techniques adopted to exploit the information potential of advanced analytical chemistry measures. In this paper, our primary focus is on elucidating the remarkable advantages of leveraging AI tools for comprehensive two-dimensional gas chromatography data (pre)processing. We illustrate how AI techniques can efficiently explore the complex datasets derived from multidimensional platforms combining comprehensive two-dimensional separations with mass spectrometry in the challenging application area of food-omics. Pattern recognition based on image processing, computer vision, and AI smelling are discussed by introducing the principles of operation, reviewing available tools and software solutions, and illustrating their potentials and limitations through selected applications. • Artificial Intelligence boost the informative potential of GC × GC. • Computer Vision based on chromatogram images. • AI smelling machine with GC × GC-MS predicts food aroma properties. • Advancing the explorative potential of comprehensive two-dimensional chromatography. • Augmented visualization combines visual features with chemical information. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) has come a long way in the last several years, both in hardware implementation and software algorithms and applications. This paper examines recent advances in artificial intelligence applications in biomedicine, such as living aid, illness diagnosis, biomedical research, and biomedical information processing. It also reviews deep learning as a technique in Artificial intelligence in healthcare and compares it to traditional methods. This review will better comprehend technology availability, keep the pace of new scientific developments, appreciate the enormous potential of artificial intelligence in biomedicine, and inspire researchers working in related domains. It is fair to say that the use of AI in biomedicine is in its early stages, like artificial intelligence itself. Artificial intelligence will continue to push the boundaries and broaden the scope of its applications as new improvements and discoveries are made, and substantial advancements are projected shortly. [ABSTRACT FROM AUTHOR]

The application of quadcopter and intelligent learning techniques in urban monitoring systems can improve flexibility and efficiency features. This paper proposes a cloud-based urban monitoring system that uses deep learning, fuzzy system, image processing, pattern recognition, and Bayesian network. The main objectives of this system are to monitor climate status, temperature, humidity, and smoke, as well as to detect fire occur-rences based on the above intelligent techniques. The quadcopter transmits sensing data of the temperature, humidity, and smoke sensors, geographical coordinates, image frames, and videos to a control station via RF communications. In the control station side, the monitoring capabilities are designed by graphical tools to show urban areas with RGB colors according to the predetermined data ranges. The evaluation process illustrates simulation results of the deep neural network applied to climate status and effects of the sensors' data changes on climate status. An illustrative example is used to draw the simulated area using RGB colors. Furthermore, circuit of the quadcopter side is designed using electric devices. [ABSTRACT FROM AUTHOR]

Heterogeneous cyberattacks against industrial control systems (ICSs) have had a strong impact on the physical world in recent decades. Connecting devices to the internet enables new attack surfaces for attackers. The intrusion of ICSs, such as the manipulation of industrial sensory or actuator data, can be the cause for anomalous ICS behaviors. This poses a threat to the infrastructure that is critical for the operation of a modern city. Nowadays, the best techniques for detecting anomalies in ICSs are based on machine learning and, more recently, deep learning. Cybersecurity in ICSs is still an emerging field, and industrial datasets that can be used to develop anomaly detection techniques are rare. In this paper, we propose an unsupervised deep learning methodology for anomaly detection in ICSs, specifically, a lightweight long short-term memory variational auto-encoder (LW-LSTM-VAE) architecture. We successfully demonstrate our solution under two ICS applications, namely, water purification and water distribution plants. Our proposed method proves to be efficient in detecting anomalies in these applications and improves upon reconstruction-based anomaly detection methods presented in previous work. For example, we successfully detected 82.16% of the anomalies in the scenario of the widely used Secure Water Treatment (SWaT) benchmark. The deep learning architecture we propose has the added advantage of being extremely lightweight. [ABSTRACT FROM AUTHOR]

Changes in the structure and function of female pelvic floor organs, muscles and surrounding connective tissue are the main causes of pelvic floor disorders (PFD), and a detailed understanding of the spatial anatomy of the pelvic floor under physiological and pathological conditions can help clinicians make accurate diagnoses and optimize treatment strategies. The widespread use of magnetic resonance imaging (MRI) technology has improved clinicians' understanding of the pelvic floor structure, but presentation of the pelvic floor structure is still limited to two-dimensional planar images and the three-dimensional (3D) spatial structure cannot be visualized. The 3D reconstruction technology based on MRI can transform the structure of the female pelvic floor organs into digital models for observation, measurement and even 3D printing for inverse analysis, which greatly promotes the clinical understanding of the pathogenesis and treatment of PFD and provides basis for individualized treatment. However, the existing MRI-based 3D reconstruction technology is not yet mature, for example different advantages and disadvantages of different commercial softwares, different algorithms of different image recognition techniques, lack of uniform evaluation criteria for 3D reconstruction models, and difficulty of balancing modeling efficiency and quality, which limit it as an effective medical research tool for PFD. This paper reviews the research of 3D digital model reconstruction of female pelvic floor tissue and organs based on MRI, aiming to provide reference for clinical diagnosis, treatment and scientific research of PFD. [ABSTRACT FROM AUTHOR]

Wafer bin maps contain vital information that helps semiconductor manufacturers to identify the root causes and defect pattern failures in wafers. Conventional manual inspection techniques in inspecting these failures are labour intensive and cause prolonged production cycle time. Therefore, automatic inspection techniques can solve this problem. This paper proposes a deep learning approach based on deep convolutional generative adversarial network (DCGAN) and a new Capsule Network (WaferCaps). DCGAN was used to upsample the original dataset and therefore increase the data used for training and balance the classes at the same time. While WaferCaps was proposed to classify the defect patterns according to eight classes. The performance of our proposed DCGAN and WaferCaps was compared with different deep learning models such as the original Capsule Network (CapsNet), CNN, and MLP. In all of our experiment, WM-811K dataset was used for the data upsampling and training. The proposed approach has shown an effective performance in generating new synthetic data and classify them with training accuracy of 99.59%, validation accuracy of 97.53% and test accuracy of 91.4%. [ABSTRACT FROM AUTHOR]

Network embedding aims to learn a low-dimensional vector for each node in networks, which is effective in a variety of applications such as network reconstruction and community detection. However, the majority of the existing network embedding methods merely exploit the network structure and ignore the rich node attributes, which tend to generate sub-optimal network representation. To learn more desired network representation, diverse information of networks should be exploited. In this paper, we develop a novel deep autoencoder framework to fuse topological structure and node attributes named FSADA. We firstly design a multi-layer autoencoder which consists of multiple non-linear functions to capture and preserve the highly non-linear network structure and node attribute information. Particularly, we adopt a pre-processing procedure to pre-process the original information, which can better facilitate to extract the intrinsic correlations between topological structure and node attributes. In addition, we design an enhancement module that combines topology and node attribute similarity to construct pairwise constraints on nodes, and then a graph regularization is introduced into the framework to enhance the representation in the latent space. Our extensive experimental evaluations demonstrate the superior performance of the proposed method. [ABSTRACT FROM AUTHOR]

In this paper an example of Advance Driving Assistance Systems (ADAS) from automotive are implemented on a small-scale vehicle. The ADAS technologies reefer to integration of multiple electronic systems of sensors and microcontrollers for detecting nearby obstacles or driver errors and respond accordingly. Several functions of ADAS uses cameras to record external environmental images for traffic signs recognition and lane assist. For implementing the above functionalities, modern software techniques are considered for processing the data acquired by the Raspberry camera on the mobile platform. Deep learning and image processing algorithms have been interconnected in a Raspberry Pi controller on the vehicle. Software development and implementation of the above algorithms have been carried out with Phyton programing language. The proposed algorithm is used to calculate the angle of turn required to stay in the current lane while reading the traffic signs for adapting the speed. Results and conclusions are presented at the end of this article. [ABSTRACT FROM AUTHOR]

Studies recently accomplished on the Enteric Nervous System have shown that chronic degenerative diseases affect the Enteric Glial Cells (EGC) and, thus, the development of recognition methods able to identify whether or not the EGC are affected by these type of diseases may be helpful in its diagnoses. In this work, we propose the use of pattern recognition and machine learning techniques to evaluate if a given animal EGC image was obtained from a healthy individual or one affect by a chronic degenerative disease. In the proposed approach, we have performed the classification task with handcrafted features and deep learning-based techniques, also known as non-handcrafted features. The handcrafted features were obtained from the textural content of the ECG images using texture descriptors, such as the Local Binary Pattern (LBP). Moreover, the representation learning techniques employed in the approach are based on different Convolutional Neural Network (CNN) architectures, such as AlexNet and VGG16, with and without transfer learning. The complementarity between the handcrafted and non-handcrafted features was also evaluated with late fusion techniques. The datasets of EGC images used in the experiments, which are also contributions of this paper, are composed of three different chronic degenerative diseases: Cancer, Diabetes Mellitus, and Rheumatoid Arthritis. The experimental results, supported by statistical analysis, show that the proposed approach can distinguish healthy cells from the sick ones with a recognition rate of 89.30% (Rheumatoid Arthritis), 98.45% (Cancer), and 95.13% (Diabetes Mellitus), being achieved by combining classifiers obtained on both feature scenarios. [ABSTRACT FROM AUTHOR]

• Developing an under-clustering algorithm to extract the most possible clusters. • Involving heuristics and deep models in the framework for clustering effectiveness. • Making up the deep learning research gap in the field of plant disease annotation. Current annotation for plant disease images depends on manual sorting and handcrafted features by agricultural experts, which is time-consuming and labour-intensive. In this paper, we propose a self-supervised clustering framework for grouping plant disease images based on the vulnerability of Kernel K-means. The main idea is to establish a cross iterative under-clustering algorithm based on Kernel K-means to produce the pseudo-labeled training set and a chaotic cluster to be further classified by a deep learning module. In order to verify the effectiveness of our proposed framework, we conduct extensive experiments on three different plant disease datatsets with five plants and 17 plant diseases. The experimental results show the high superiority of our method to do image-based plant disease classification over balanced and unbalanced datasets by comparing with five state-of-the-art existing works in terms of different metrics. [ABSTRACT FROM AUTHOR]

In this paper, we consider a palm print characteristic which has taken wide attentions in recent studies. We focused on palm print verification problem by designing a deep network called a palm convolutional neural network (PCNN). This network is adapted to deal with two-dimensional palm print images. It is carefully designed and implemented for palm print data. Palm prints from the Hong Kong Polytechnic University Contact-free (PolyUC) 3D/2D hand images dataset are applied and evaluated. The results have reached the accuracy of 97.67%, this performance is superior and it shows that our proposed method is efficient. [ABSTRACT FROM AUTHOR]

The writing style of the same writer varies from instance to instance in Arabic and English handwritten digit recognition, making handwritten digit recognition challenging. Currently, deep learning approaches are applied in many applications, including convolutional neural networks (CNNs) modified to produce other models, such as local binary convolutional neural networks (LBCNNs). An LBCNN is created by fusing a local binary pattern (LBP) with a CNN by reformulating the LBP as a convolution layer called a local binary convolution (LBC). However, LBCNNs suffer from the random assignment of 1, 0, or -1 to LBC weights, making LBCNNs less robust. Nevertheless, using another LBP-based technique, such as center-symmetric local binary patterns (CS-LBPs), can address such issues. In this paper, a new model based on CS-LBPs is proposed called center-symmetric local binary convolutional neural networks (CS-LBCNN), which addresses the issues of LBCNNs. Furthermore, an enhanced version of CS-LBCNNs called threshold center-symmetric local binary convolutional neural networks (TCS-LBCNNs) is proposed, which addresses another issue related to the zero-thresholding function. Finally, the proposed models are compared to state-of-the-art models, proving their ability by producing a more accurate and significant classification rate than the existing LBCNN models. For the bilingual dataset, the TCS-LBCNN enhances the accuracy of the LBCNN and CS-LBCNN by 0.15% and 0.03%, respectively. In addition, the comparison shows that the accuracy acquired by the TCS-LBCNN is the second-highest using the MNIST and MADBase datasets. [ABSTRACT FROM AUTHOR]

Modern power systems are experiencing the challenge of high uncertainty with the increasing penetration of renewable energy resources and the electrification of heating systems. In this paradigm shift, understanding electricity users' demand is of utmost value to the retailers, aggregators, and policymakers. However, behind-the-meter (BTM) equipment and appliances at the household level are unknown to the other stakeholders mainly due to privacy concerns and tight regulations. In this paper, we seek to identify residential consumers based on their BTM equipment, mainly rooftop photovoltaic (PV) systems and electric heating, using imported/purchased energy data from utility meters. To solve this problem with an interpretable, fast, secure, and maintainable solution, we propose an integrated method called Interpretable Refined Motifs And binary Classification (IRMAC). The proposed method comprises a novel shape-based pattern extraction technique, called Refined Motif (RM) discovery, and a single-neuron classifier. The first part extracts a sub-pattern from the long time series considering the frequency of occurrences, average dissimilarity, and time dynamics while emphasising specific times with annotated distances. The second part identifies users' types with linear complexity while preserving the transparency of the algorithms. With the real data from Australia and Denmark, the proposed method is tested and verified in identifying PV owners and electrical heating system users. The performance of the IRMAC is studied and compared with various state-of-the-art methods. The proposed method reached an accuracy of 96% in identifying rooftop PV users and 94.4% in identifying electric heating users, which is comparable to the best solution based on deep learning, while the speed of the inference model is a thousand times faster. Last but not least, the proposed method is a transparent algorithm, which can tackle the concerns regarding the agnostic decision-making process when policies prohibit some machine learning methods. [ABSTRACT FROM AUTHOR]

• The systematic failure pattern recognition method for wafer map is explored. • A deep convolutional neural network model is established. • This model is an end-to-end 19-layer network structure. • This model can automatically extract effective classification features. • The mode can enhance the reliability and efficiency of fault pattern recognition. The objective of this paper is to propose a systematic failure pattern recognition for wafer map based on neural networks. A deep convolutional neural network (DCNN) model which includes the convolutional layer, batch normalization layer, Relu layer, maximum pooling layer, full connection layer, Softmax layer and classification layer is established for the problem of failure pattern recognition of wafer map. This model is an end-to-end 19-layer network, which can actively learn and automatically extract effective classification features. After grayscale and median filtering, the wafer map can be imported into the network for automatic failure classification without special feature extraction. On this basis, we build a dual-source DCNN structure by combining decision-level information entropy fusion. The verification results of the model in the actual wafer map database WM-811K show that the model exhibits good performance in identifying nine kinds of common failure patterns, and has advantages in identifying non-pattern wafer patterns without failure patterns. The dual-source DCNN structure has a better classification effect than the single-source DCNN structure, and the overall recognition accuracy of the dual-source DCNN structure reaches 98.34%. [ABSTRACT FROM AUTHOR]

Detection transformer (DETR) combines convolutional neural network (CNN) with transformer, providing a more advanced idea. In this paper, an object detection model based on DETR is proposed for fire and smoke detection. Compared with other methods that based on deep learning, the proposed one simplifies the pipeline of detection and builds an end-to-end detector. At the same time, the original DETR model usually requires long training time and large amount of computation, resulting in relatively poor performance in detection speed and accuracy, and it shows to be not friendly for small or early fire detection. Therefore, when designing the proposed network model, a normalization-based attention module is added in the feature extraction stage to highlight the effective features, being beneficial to the process of encoding. A multiscale deformable attention is also used in the encoder–decoder structure, accelerating therefore, the process of convergence during the training of the model, which includes the enhancement of the detection of small objects. Also, considering the cost of computation, the number of layers in the encoder–decoder structure is redefined to reduce the complexity of the model, and that also reduces the requirements of the application equipment. Detailed experiments are conducted on three self-built datasets and two public video sets. The results show that, the proposed method has an excellent performance on all of the datasets considered here. [ABSTRACT FROM AUTHOR]

Many current intrinsically interpretable machine learning models can only handle the data that are linear, low-dimensional, and relatively independent attributes and often with discrete attribute values, while the models that are capable of handling high-dimensional nonlinear data, like deep learning, have very poor interpretability. Based on the geometric characteristics, a new idea of accurately wrapping the data region with minimum-volume geometry is proposed for pattern classification. The Variance Hyper Rectangle (VHR) model presented in this paper is a realization of the idea. The VHR model uses the minimum-volume hyper rectangles, obtained through projection variance calculation, to wrap the regions occupied by a category of data, hence it has strong and clear geometric interpretability. In addition, the VHR model is well suited for large data volume, as it approaches the linear complexity in both time and space. Extensive qualitative and quantitative experiments are performed on seven real-world data sets, demonstrating that VHR outperforms the state-of-the-art interpretable methods while running quickly. [Display omitted] • VHR has strong geometric interpretability, and is much reliable and trustworthy. • VHR can provide a clear range of values in each direction for a category of data. • VHR naturally supports incremental learning without any extra processing. • VHR has great performance and stability, and is able to handle big data. [ABSTRACT FROM AUTHOR]

• An non-contact method was invented for tamping counting during the dynamic compaction construction. • The hammer motion recognition model was created based on the operating states of dynamic compaction crane (DCC). • The applicability of object detection methods based on cooperative targets or deep learning in diverse environments is discussed. Dynamic compaction method (DCM) is currently one of the most commonly used foundation reinforcement techniques. However, manual monitoring is still the mainstream way of DCM tamping counting with low efficiency and high cost. This paper focuses on the tamping times of DCM, and proposes a non-contact Intelligent Monitoring Method for Tamping Times (IM2T2) based on machine vision and pattern recognition technology. The hammer detection methods based on cooperative targets, YOLOv4 and YOLOv4-tiny are compared, and then the motion model based on the hammer position of construction image series is proposed and the vision-based full automatic measurement of tamping times is realized. Moreover, a field test was carried out to verify the applicability of above method. The results of the tamping times measurement indicate that the proposed method can measure the count of tamping under general working conditions with quite high accuracy. [ABSTRACT FROM AUTHOR]

Deep learning (DL) has achieved significant attention in the field of infrared (IR) and visible (VI) image fusion, and several attempts have been made to enhance the quality of the final fused image. It produces better results than conventional methods; however, the captured image cannot acquire useful information due to environments with poor lighting, fog, dense smoke, haze, and the noise generated by sensors. This paper proposes an adaptive fuzzy-based preprocessing enhancement method that automatically enhances the contrast of images with adaptive parameter calculation. The enhanced images are then decomposed into base and detail layers by anisotropic diffusion-based edge-preserving filters that remove noise while smoothing the edges. The detailed parts are fed into four convolutional layers of the VGG-19 network through transfer learning to extract features maps. These features maps are fused by multiple fusion strategies to obtain the final fused detailed layer. The base parts are fused by the PCA method to preserve the energy information. Experimental results reveal that our proposed method achieves state-of-the-art performance compared with existing fusion methods in a subjective evaluation through the visual experience of experts and statistical tests. Moreover, the objective assessment parameters are conducted by various parameters ( F M I , S S I M a , A P I , EN , Q F AB , and N F A B ) which were used in the comparison method. The proposed method achieves 0.2651 to 0.3951, 0.5827 to 0.8469, 56.3710 to 71.9081, 4.0117 to 7.9907, and 0.6538 to 0.8727 gain for F M I , S S I M a , A P I , EN , and Q F AB , respectively. At the same time, the proposed method has more noise reduction (0.3049 to 0.0021) that further justifies the efficacy of the proposed method than conventional methods. [ABSTRACT FROM AUTHOR]

In the current pandemic, smart technologies such as cognitive computing, artificial intelligence, pattern recognition, chatbot, wearables, and blockchain can sufficiently support the collection, analysis, and processing of medical data for decision making. Particularly, to aid medical professionals in the disease diagnosis process, cognitive computing is helpful by processing massive quantities of data rapidly and generating customized smart recommendations. On the other hand, the present world is facing a pandemic of COVID-19 and an earlier detection process is essential to reduce the mortality rate. Deep learning (DL) models are useful in assisting radiologists to investigate the large quantity of chest X-ray images. However, they require a large amount of training data and it needs to be centralized for processing. Therefore, federated learning (FL) concept can be used to generate a shared model with no use of local data for DL-based COVID-19 detection. In this view, this paper presents a federated deep learning-based COVID-19 (FDL-COVID) detection model on an IoT-enabled edge computing environment. Primarily, the IoT devices capture the patient data, and then the DL model is designed using the SqueezeNet model. The IoT devices upload the encrypted variables into the cloud server which then performs FL on major variables using the SqueezeNet model to produce a global cloud model. Moreover, the glowworm swarm optimization algorithm is utilized to optimally tune the hyperparameters involved in the SqueezeNet architecture. A wide range of experiments were conducted on benchmark CXR dataset, and the outcomes are assessed with respect to different measures . The experimental outcomes pointed out the enhanced performance of the FDL-COVID technique over the other methods. [ABSTRACT FROM AUTHOR]

• Deep learning concepts borrowed for building a powerful feature extraction framework. • A novel recurrent spatial-temporal fusion demonstrates a significant performance. • A bi-directional framework is also proposed equipped with simple TD features. • Results are benchmarked against several algorithms across 82 subjects. • RSTF significantly outperforms all other methods, including LSTM and BiLSTM. Current state-of-the-art myoelectric interfaces employ traditional pattern recognition (PR) algorithms to decode the Electromyogram (EMG) signals into hand movements for controlling artificial limbs. Recently, deep learning (DL) models have also been exploited for EMG feature learning/extraction. Models like Convolutional Neural Networks (CNN), which capture the spatial correlations, and Long Short-Term Memory (LSTM), which capture the non-linear temporal dynamics of EMG time-series data, have been shown to outperform traditional EMG PR systems. Nevertheless, the large number of model parameters, long training times, and large amounts of data required to train these DL models remain limiting factors that may hinder their translation into clinically viable prostheses. Consequently, rather than applying DL directly, this paper leverages concepts derived from these models to build upon our proposed concept of a Fusion of Time Domain Descriptors (FTDD). The FTDD are augmented with Range Spatial Filtering (RSF) to capture the spatial correlations and combined into an LSTM-style framework. This process, denoted as Recurrent Spatial-Temporal Fusion (RSTF), can be applied in combination with any traditional feature extraction method to exploit temporal and spatial correlations, with the potential for bi-directional applications. The advantages of the proposed RSTF method include (1) the memory concept, capturing long- and short-term spatial and temporal dependencies of the EMG signals, (2) significantly improved performance outperforming other state-of-the-art models and (3) the simplicity and the fairly low computational costs for feature extraction. Results are bench-marked against several feature extraction methods, proving the power of the RSTF using data from 82 subjects from five EMG databases with varying recording characteristics. The proposed method significantly outperforms all other methods tested for EMG pattern recognition, including a deep LSTM and other CNN methods previously reported in the literature and at a fracture of the computational cost. On the most challenging dataset, improvements of as much as 15% were found. [ABSTRACT FROM AUTHOR]