Search

Financial risk warning is a crucial technical issue for enterprises. Traditionally, it relied on modeling analysis of a single data type, which does not fully capture the diverse characteristics of financial risk activities. To address this, this paper introduces a multimodal deep learning framework driven by semantic analysis and image processing for intelligent warning of enterprise financial risks. Initially, natural language processing algorithms analyze textual data such as financial statements, news reports, and social media comments within the industry. Concurrently, two convolutional neural network models, M-CNN and M-RNN, extract features from images and chart data. These textual and visual feature representations are then fused to create a multimodal deep neural network framework. Extensive experimental evaluation and comparative analysis of the proposed framework were conducted. The results indicate that the financial risk rate of consumer fraud graph analysis varies significantly, displaying a fluctuating state with values ranging from 2.1% to 16.8%. Compared to other methods, the proposed approach demonstrates superior performance in financial risk warning tasks, with the risk rate increasing from 1.2% to 26.5% as it iterates from 1 to 6. [ABSTRACT FROM AUTHOR]

This paper presents a novel use of a bio-inspired retina model as a scene preprocessing stage for the recognition of letters and numbers on Chilean vehicle license plates. The goal is to improve the effectiveness and ease of pattern recognition. Inspired by the responses of mammalian retinas, this retinal model reproduces both the natural adjustment of contrast and the enhancement of object contours by parvocellular cells. Among other contributions, this paper provides an in-depth exploration of the architecture, advantages, and limitations of the model; investigates the tuning parameters of the model; and evaluates its performance when integrating a convolutional neural network and a spiking neural network into an optical character recognition (OCR) algorithm, using 40 different genuine license plate images as a case study and for testing. The results obtained demonstrate the reduction of error rates in character recognition based on convolutional neural networks (CNNs), spiking neural networks (SNNs), and OCR. It is concluded that this bio-inspired retina model offers a wide spectrum of potential applications to further explore, including motion detection, pattern recognition, and improvement of dynamic range in images, among others. [ABSTRACT FROM AUTHOR]

Low‐light image enhancement is a longstanding challenge in low‐level vision, as images captured in low‐light conditions often suffer from significant aesthetic quality flaws. Recent methods based on deep neural networks have made impressive progress in this area. In contrast to mainstream convolutional neural network (CNN)‐based methods, an effective solution inspired by the transformer, which has shown impressive performance in various tasks, is proposed. This solution is centred around two key components. The first is an image synthesis pipeline, and the second is a powerful transformer‐based pre‐trained model, known as the low‐light image enhancement transformer (LIET). The image synthesis pipeline includes illumination simulation and realistic noise simulation, enabling the generation of more life‐like low‐light images to overcome the issue of data scarcity. LIET combines streamlined CNN‐based encoder‐decoders with a transformer body, efficiently extracting global and local contextual features at a relatively low computational cost. The extensive experiments show that this approach is highly competitive with current state‐of‐the‐art methods. The codes have been released and are available at LIET. [ABSTRACT FROM AUTHOR]

Decomposing an image into meaningful components is an effective way for image processing. Making full use of cartoon features and texture features with deep neural networks may further achieve the stable noise cancellation performance for low-dose computed tomography (LDCT) denoising. However, the performance of conventional deep neural networks is limited by their black-box nature. To overcome this drawback, we proposed an algorithm unrolling approach for LDCT denoising, namely, the cartoon-texture features guided network (CTFG-Net). The CTFG-Net integrates a cartoon sub-network, a texture sub-network, and a fusion sub-network into an end-to-end convolutional neural network model. A cartoon decomposition algorithm and a texture decomposition algorithm were unrolled to the cartoon sub-network and the texture sub-network, respectively. These two sub-networks were employed to decompose a LDCT image into its cartoon components and texture components. During the process of image decomposition, these two sub-networks exploited cartoon features and texture features to greatly suppress noise and remove artifacts in the corresponding components. At last, the final denoised results were yielded through a data fusion of the denoised cartoon components and the denoised texture components. The proposed model was tested with three datasets, Mayo dataset, Piglet dataset, and AbdomenCT-1K dataset. Qualitative results show that the CTFG-Net achieved better denoising effects for LDCT images than the competing algorithms. Compared to the competing algorithms, the proposed CTFG-Net also improved the peak signal-to-noise ratio, structural similarity index measure, and visual information fidelity by (0.3994 dB and 0.7692 dB), (0.0115 and 0.0056), and (0.0751 and 0.0433) with Mayo dataset and Piglet dataset. [ABSTRACT FROM AUTHOR]

Image processing with the help of machine learning algorithms is breaking barriers in various fields of study, especially in the medical field. Deep learning image classification algorithms have made disease detection based on images in an easy manner thus assists the medical professional to taking quick decisions. This paper discusses various deep learning algorithms and transfer learning methods used to classify various eye diseases. Kaggle Eye dataset is used for this purpose. We compared four deep learning algorithms, namely EfficientNetB3, Inception V3, VGG 19 and Convolutional Neural Network models. Various categories of eye diseases, namely Cataract, Diabetic Retinopathy, Glaucoma are considered for classification with normal eye with the help of the scanned images. The strengths and weaknesses of these models are compared based on Precision, Recall, Accuracy and F1 score. In an identical testing environment, EfficientNet B3 outperforms the other algorithms and provides better accuracy for the classification of eye diseases. [ABSTRACT FROM AUTHOR]

Corn is rich in fibre, vitamins, and minerals, and it is a nutritious source of carbohydrates. The area under corn cultivation is very large because, in addition to providing food for humans and animals, it is also used for raw materials for industrial products. Corn cultivation is exposed to the damage of various pests such as armyworm. A regional monitoring of pests is intended to actively track the population of this pest in a specific geography; one of the ways of monitoring is using the image processing technology. Therefore, the aim of this research was to identify healthy and armyworm-infected leaves using image processing and deep neural network in the form of 4 structures named AlexNet, DenseNet, EfficientNet, and GoogleNet. A total of 4500 images, including healthy and infected leaves, were collected. Next, models were trained by train data. Then, test data were evaluated using the evaluation criteria such as accuracy, precision, and F score. Results indicated all the classifiers obtained the precision above 98%, but the EfficientNet-based classifier was more successful in classification with the precision of 100%, accuracy of 99.70%, and F-score of 99.68%. [ABSTRACT FROM AUTHOR]

In order to balance accuracy and real‐time performance in semantic segmentation, this paper proposes a real‐time semantic segmentation algorithm model based on attention mechanism and multi‐branch feature fusion using Fast convolutional neural network model (Fast‐SCNN). In this method, the spatial detail feature enhancement branch is introduced to enhance spatial detail features firstly. Then, through rational design of fusion module, the feature information of each branch is optimized to achieve better fusion of deep and shallow features. At the end of the feature fusion module, an adaptive feature enhancement focus module is introduced to capture the interdependence between remote pixels. The experimental results show that the proposed algorithm achieves 71.55% segmentation accuracy on Cityscapes dataset, the reasoning speed FPS is 97.6 frames/s, and the number of parameters is 1.39 M, which verifies the effectiveness of the network model constructed by the algorithm. Code is available at https://github.com/ccchhheeennn/model. [ABSTRACT FROM AUTHOR]

Background: Artificial intelligence (AI) has shown great promise in the field of healthcare as a means of improving the diagnosis of skin cancer. The objective of this research is to enhance the precision and effectiveness of skin cancer identification by the incorporation of convolutional neural networks (CNNs) and discrete wavelet transformation (DWT). Making use of AI-driven techniques has the potential to completely transform the diagnosis process by providing quicker and more accurate evaluations of skin lesions. In an effort to improve dermatology and give physicians reliable resources for early and precise skin cancer diagnosis, this work explores the combination of CNNs with DWT. Methods: The accurate and timely classification of skin cancer lesions plays a crucial role in early diagnosis and effective treatment. In this, we propose a novel approach for skin cancer classification using discrete wavelet transformation (DWT). The DWT is employed to extract relevant features from skin lesion images, which are then used to train a classification model. The effectiveness of the suggested approach is assessed through the examination of a dataset of skin lesion images with known classes (malignant or benign). Results: The outcomes of the experiment demonstrate that the suggested model successfully attained a classification result of sensitivity as 94% and specificity as 91% when compared with artificial neural network (ANN) and multilayer perceptron methods. Conclusions: The HAM 10000 dataset is employed to explore and evaluate the effectiveness of the proposed model, leading to improved accuracy compared to the existing machine learning algorithms in utilization. The results demonstrate the effectiveness of the DWT-based approach in accurately classifying skin cancer lesions, thus aiding in early detection and diagnosis. [ABSTRACT FROM AUTHOR]

This study aims to use a machine learning (ML) model to accurately classify four datasets of cotton crop leaves as either infected or healthy. Bacterial blight, Curly virus, Fussarium Wilt, and healthy leaves were used as the datasets for the study. ML is a useful tool in detecting cotton leaf diseases and can minimize the rate of disease. The problem is that without machine learning technique it is very difficult and time consuming to detect the diseases then to sort out this problem a machine learning model is proposed and to test the accuracy of the proposed model, the confusion matrix concept was used. The researchers have done their research works to diagnose the diseases by using (ML) model but the drawback of their research was that the results which were given by the different (ML) models were not accurate. The target of the study was to identify diseases affecting the cotton plant in the early stages using traditional techniques. However, utilizing various image processing techniques and machine learning algorithms, including a convolutional neural network, proved to be helpful in diagnosing the diseases. This technological approach can simplify the detection of damaged leaves and minimize the efforts of farmers in detecting those diseases. Cotton is a natural fiber produced on a large scale, and it is grown on 2.5% of overall agronomic land. The detection of cotton leaf diseases is crucial to maintain the crop's productivity and provide reliable earnings to farmers. A confusion matrix is N X N matrix used for evaluating the performance of a classification model, where N is the number of target classes. The matrix compares the actual target values with those predicted by machine learning model. This technique has four parameters to test the accuracy of the results which is given in my research work. [ABSTRACT FROM AUTHOR]

This document is a summary of a special issue titled "Deep and Machine Learning for Image Processing: Medical and Non-medical Applications." The special issue explores the use of machine learning and deep learning technologies in image processing, with a focus on medical and non-medical applications. The contributions cover a wide range of topics, including the use of ensemble models for diagnosing myocardial infarction, camouflaged object detection, crime analysis using seasonal autoregressive integrated moving average models, monkeypox diagnosis using metaheuristic optimization algorithms, instance segmentation for power operation monitoring, long-distance person detection, hate-speech detection on social media, perceptual image quality prediction, material discrimination using colorization, compressed sensing magnetic resonance imaging, quantization for low-power accelerators, and computer-aided diagnosis for skin diseases. The document also highlights future research directions, such as refining integrated ML/DL models, broadening predictive modeling, enhancing data fusion, advancing imaging and material discrimination techniques, developing context-aware natural language processing models, improving object detection, optimizing for edge computing, innovating in medical imaging, and assessing multimedia content quality. The special issue aims to showcase the current state of research in ML and DL applications for image processing and inspire further advancements in the field. [Extracted from the article]

One of the major threats in computer vison and image processing areas is image tarnishing due to noisy pixels. Image tarnishing occurs when clean images are subjected to multiple feature changes, length transmission, or due to improper image capturing. The growth of Autoencoders in Artificial Neural Networks has paved way to efficient noise removal. In this work, an architecture oriented analysis of stacked denoising autoencoders is implemented by considering two perspectives of noise, namely noise as an inverse problem and noise as a residual problem. Autoencoders have been modelled by considering the approaches: Shallow Model, Deep-Series and Parallel Series models. Autoencoders with Skip Connections and Denoising CNNs have also been implemented and analyzed in order to propose an efficient, optimized and best working model for denoising. The dataset used is Common Object in Context (COCO) and the performance of the model is evaluated based on Structural Similarity Index (SSIM) and Peak Signal to Noise Ratio (PSNR). The analysis has presented the impact of the model structures of the autoencoders, which implicitly tend to treat noise as an inverse problem. [ABSTRACT FROM AUTHOR]

There are several image inverse tasks, such as inpainting or super-resolution, which can be solved using deep internal learning, a paradigm that involves employing deep neural networks to find a solution by learning from the sample itself rather than a dataset. For example, Deep Image Prior is a technique based on fitting a convolutional neural network to output the known parts of the image (such as non-inpainted regions or a low-resolution version of the image). However, this approach is not well adjusted for samples composed of multiple modalities. In some domains, such as satellite image processing, accommodating multi-modal representations could be beneficial or even essential. In this work, Multi-Modal Convolutional Parameterisation Network (MCPN) is proposed, where a convolutional neural network approximates shared information between multiple modes by combining a core shared network with modality-specific head networks. The results demonstrate that these approaches can significantly outperform the single-mode adoption of a convolutional parameterisation network on guided image inverse problems of inpainting and super-resolution. [ABSTRACT FROM AUTHOR]

With the extensive adoption of generative models across various domains, the protection of copyright for these models has become increasingly vital. Some researchers suggest embedding watermarks in the images generated by these models as a means of preserving IP rights. In this paper, we find that existing generative model watermarking introduces high-frequency artifacts in the high-frequency spectrum of the marked images, thereby compromising the imperceptibility and security of the generative model watermarking system. Given this revelation, we propose an innovative image watermarking technology that takes into account frequency-domain imperceptibility. Our approach abandons the conventional convolutional neural network (CNN) structure typically used as the watermarking embedding network in popular watermarking techniques. This helps the image watermarking system to avoid the inherent high-frequency artifacts commonly produced by CNNs. In addition, we design a frequency perturbation generation network to generate low-frequency perturbations. These perturbations are subsequently added as watermarks to the low-frequency components of the carrier image, thus minimizing the impact of the watermark embedding process on the high-frequency properties of the image. The results show that our proposed watermarking framework can effectively embed low-frequency perturbation watermarks into images and effectively suppress high-frequency artifacts in images, thus significantly improving the frequency-domain imperceptibility and security of the image watermarking system. The introduced approach enhances the average invisibility performance in the frequency domain by up to 24.9 % when contrasted with prior methods. Moreover, the method attains superior image quality (>50 dB) in the spatial domain, accompanied by a 100 % success rate in watermark extraction in the absence of attacks. This underscores its capability to uphold the efficacy of the protected network and preserve the integrity of the watermarking process. It always maintains excellent imperceptibility and robustness. Thus, the framework shows great potential as a state-of-the-art solution for protecting intellectual property. [ABSTRACT FROM AUTHOR]

The rapid development of neural network technology promotes the neural network model driven by big data to overcome the texture effect of complex objects. Due to the limitations in complex scenes, it is necessary to establish custom template matching and apply it to the research of many fields of computational vision technology. The dependence on high-quality small label sample database data is not very strong, and the machine learning system of deep feature connection to complete the task of texture effect inference and speculation is relatively poor. The style transfer algorithm based on neural network collects and preserves the data of patterns, extracts and modernizes their features. Through the algorithm model, it is easier to present the texture color of patterns and display them digitally. In this paper, according to the texture effect reasoning of custom template matching, the 3D visualization of the target is transformed into a 3D model. The high similarity between the scene to be inferred and the user-defined template is calculated by the user-defined template of the multi-dimensional external feature label. The convolutional neural network is adopted to optimize the external area of the object to improve the sampling quality and computational performance of the sample pyramid structure. The results indicate that the proposed algorithm can accurately capture the significant target, achieve more ablation noise, and improve the visualization results. The proposed deep convolutional neural network optimization algorithm has good rapidity, data accuracy and robustness. The proposed algorithm can adapt to the calculation of more task scenes, display the redundant vision-related information of image conversion, enhance the powerful computing power, and further improve the computational efficiency and accuracy of convolutional networks, which has a high research significance for the study of image information conversion. [ABSTRACT FROM AUTHOR]

How to represent a human face pattern? While it is presented in a continuous way in human visual system, computers often store and process it in a discrete manner with 2D arrays of pixels. The authors attempt to learn a continuous surface representation for face image with explicit function. First, an explicit model (EmFace) for human face representation is proposed in the form of a finite sum of mathematical terms, where each term is an analytic function element. Further, to estimate the unknown parameters of EmFace, a novel neural network, EmNet, is designed with an encoder‐decoder structure and trained from massive face images, where the encoder is defined by a deep convolutional neural network and the decoder is an explicit mathematical expression of EmFace. The authors demonstrate that our EmFace represents face image more accurate than the comparison method, with an average mean square error of 0.000888, 0.000936, 0.000953 on LFW, IARPA Janus Benchmark‐B, and IJB‐C datasets. Visualisation results show that, EmFace has a higher representation performance on faces with various expressions, postures, and other factors. Furthermore, EmFace achieves reasonable performance on several face image processing tasks, including face image restoration, denoising, and transformation. [ABSTRACT FROM AUTHOR]

In recent years, with the growing proliferation of photovoltaics (PV), accurate nowcasting of PV power has emerged as a challenge. Global horizontal irradiance (GHI), which is a key factor influencing PV power, is known to be highly variable as it is determined by short-term meteorological phenomena, particularly cloud movement. Deep learning and computer vision techniques applied to all-sky imagery are demonstrated to be highly accurate nowcasting methods, as they encode crucial information about the sky's state. While these methods utilize deep neural network models, such as Convolutional Neural Networks (CNN), and attain high levels of accuracy, the training of image-based deep learning models demands significant computational resources. In this work, we present a computationally economical estimation technique, based on a deep learning model. We utilize both all-sky imagery and meteorological data, however, information on the sky's state is encoded as a feature vector extracted using traditional image processing methods. We introduce six all-sky image features utilizing detailed knowledge of meteorological and physical phenomena, significantly decreasing the amount of input data and model complexity. We investigate the accuracy of the determined global and diffuse radiation for different combinations of meteorological parameters. The model is evaluated using two years of measurements from an on-site all-sky camera and an adjacent meteorological station. Our findings demonstrate that the model provides comparable accuracy to CNN-based methods, yet at a significantly lower computational cost. [ABSTRACT FROM AUTHOR]

The presented paper investigates the problem of endoscopic bleeding detection in endoscopic videos in the form of a binary image classification task. A set of definitions of high-level visual features of endoscopic bleeding is introduced, which incorporates domain knowledge from the field. The high-level features are coupled with respective feature descriptors, enabling automatic capture of the features using image processing methods. Each of the proposed feature descriptors outputs a feature activation map in the form of a grayscale image. Acquired feature maps can be appended in a straightforward way to the original color channels of the input image and passed to the input of a convolutional neural network during the training and inference steps. An experimental evaluation is conducted to compare the classification ROC AUC of feature-extended convolutional neural network models with baseline models using regular color image inputs. The advantage of feature-extended models is demonstrated for the Resnet and VGG convolutional neural network architectures. [ABSTRACT FROM AUTHOR]

Advances in technology have led to advances in breast cancer screening by detecting symptoms that doctors have overlooked. In this paper, an automatic detection system for breast cancer cases based on Internet of Things (IoT) is proposed. First, using IoT technology, direct medical images are sent to the data repository after the suspicious person's visit through medical equipment equipped with IoT. Then, in order to help radiologists, interpret medical images as best as possible, we use four pre-trained convolutional neural network models including InceptionResNetV2, InceptionV3, VGG19 and ResNet152. These models are combined by an ensemble classifier. Also, these models are used to accurately predict cases with breast cancer, healthy people, and cases with pneumonia by using two datasets of X-RAY and CT-scan in a three-class classification. Finally, the best result obtained for CTscan images belongs to InceptionResNetV2 architecture with 99.36% accuracy and for X-RAY images belongs to InceptionV3 architecture with 96.94% accuracy. The results show that this method leads to a reduction in daily visits to medical centers and thus reduces the pressure on the medical care system. It also helps radiologists and medical staff to detect breast cancer in its early stages. [ABSTRACT FROM AUTHOR]

Catastrophic forgetting, which means a rapid forgetting of learned representations while learning new data/samples, is one of the main problems of deep neural networks. In this paper, we propose a novel incremental learning framework that can address the forgetting problem by learning new incoming data in an online manner. We develop a new incremental learning framework that can learn extra data or new classes with less catastrophic forgetting. We adopt the hippocampal memory process to the deep neural networks by defining the effective maximum of neural activation and its boundary to represent a feature distribution. In addition, we incorporate incremental QR factorization into the deep neural networks to learn new data with both existing labels and new labels with less forgetting. The QR factorization can provide the accurate subspace prior, and incremental QR factorization can reasonably express the collaboration between new data with both existing classes and new class with less forgetting. In our framework, a set of appropriate features (i.e., nodes) provides improved representation for each class. We apply our method to the convolutional neural network (CNN) for learning Cifar-100 and Cifar-10 datasets. The experimental results show that the proposed method efficiently alleviates the stability and plasticity dilemma in the deep neural networks by providing the performance stability of a trained network while effectively learning unseen data and additional new classes. [ABSTRACT FROM AUTHOR]

Security systems are based on the monitoring of specific areas of the facility. The cameras record the selected place for the whole day. Unfortunately, it is difficult to automatically analyze the recorded situations mainly through manual analysis. In this paper, we propose an innovative automatic monitoring data analysis system. To minimize the amount of processed data, a heuristic-based method is proposed to analyze frames. Heuristic algorithm is adapted to image analysis. If the algorithm detects significant changes in pixel values, the frame is sent further to the convolutional neural network. The proposed solution is based on centralized federated learning that allows training a shared model on a local dataset. This guarantees surveillance recordings' privacy with a shared model. The proposal is a hybrid solution that was presented as a mathematical model, tested and compared with other known solutions. Based on conducted experiments, the proposed image processing system minimizes the number of calculations by the hybrid approach that can be valuable for IoT applications. The proposed solution is more effective than the existing solution due to the use of classifiers for the analysis of single frames. • Automatic analysis of video frames through the use of a heuristic algorithm. • Analysis of selected video frames through the use of deep neural networks. • A decision module for the analyzed data and operation automation. • The system is based on individual implementation with a federated learning approach. [ABSTRACT FROM AUTHOR]

The paper considers a class of discrete-time cellular neural networks (DT-CNNs) obtained by applying Euler's discretization scheme to standard CNNs. Let T be the DT-CNN interconnection matrix which is defined by the feedback cloning template. The paper shows that a DT-CNN is convergent, i.e. each solution tends to an equilibrium point, when T is symmetric and, in the case where T + E n is not positive-semidefinite, the step size of Euler's discretization scheme does not exceed a given bound ( E n is the n × n unit matrix). It is shown that two relevant properties hold as a consequence of the local and space-invariant interconnecting structure of a DT-CNN, namely: (1) the bound on the step size can be easily estimated via the elements of the DT-CNN feedback cloning template only; (2) the bound is independent of the DT-CNN dimension. These two properties make DT-CNNs very effective in view of computer simulations and for the practical applications to high-dimensional processing tasks. The obtained results are proved via Lyapunov approach and LaSalle's Invariance Principle in combination with some fundamental inequalities enjoyed by the projection operator on a convex set. The results are compared with previous ones in the literature on the convergence of DT-CNNs and also with those obtained for different neural network models as the Brain-State-in-a-Box model. Finally, the results on convergence are illustrated via the application to some relevant 2D and 1D DT-CNNs for image processing tasks. [ABSTRACT FROM AUTHOR]

To capture the real-world luminance values, High Dynamic Range (HDR) image processing has been developed. HDR images have a richer content than the widely-used Standard Dynamic Range (SDR) images, and are used in a number of situations, e.g. in film industry. As HDR displays are more and more commercially available, we need to be able to process HDR images as well as SDR ones (for example, devising denoising algorithms, inpainting or anti-aliasing). The most powerful methods to process images are now deep neural networks. However, the training of such networks calls for a lot of images, and HDR images datasets are relatively small. One way to generate HDR images is inverse tone mapping operators (iTMOs). They are algorithms that expand the dynamic range of SDR images. In this paper, we propose HDR-LFNet, a novel iTMO, and its HDR training dataset. Our method merges several existing handcrafted iTMOs, combined in a supervised neural network to produce an HDR output. Our lightweight network requires less training images than state-of-the-art methods, and has faster training phase. Besides, the quality of the generated images is similar to the state-of-the-art. We present the architecture as well as the subjective and experimental evaluations of our method. [Display omitted] • We propose a new Inverse Tone Mapping Operator to create High Dynamic Range images. • Convolutional Neural Network are used to fuse together existing methods. • We performed a subjective study to evaluate our method. • The network's weights and training set are available online. [ABSTRACT FROM AUTHOR]

We increasingly rely on deep learning algorithms to process colossal amount of unstructured visual data. Commonly, these deep learning algorithms are deployed as software models on digital hardware, predominantly in data centers. Intrinsic high energy consumption of Cloud-based deployment of deep neural networks (DNNs) inspired researchers to look for alternatives, resulting in a high interest in Spiking Neural Networks (SNNs) and dedicated mixed-signal neuromorphic hardware. As a result, there is an emerging challenge to transfer DNN architecture functionality to energy-efficient spiking non-volatile memory (NVM)-based hardware withminimal loss in the accuracy of visual data processing. Convolutional Neural Network (CNN) is the staple choice of DNN for visual data processing. However, the lack of analog-friendly spiking implementations and alternatives for some core CNN functions, such as MaxPool, hinders the conversion of CNNs into the spike domain, thus hampering neuromorphic hardware development. To address this gap, in this work, we propose MaxPool with temporal multiplexing for Spiking CNNs (SCNNs), which is amenable for implementation in mixed-signal circuits. In this work, we leverage the temporal dynamics of internal membrane potential of Integrate & Fire neurons to enable MaxPool decision-making in the spiking domain. The proposed MaxPool models are implemented and tested within the SCNNarchitecture using amodified version of the aihwkit framework, a PyTorch-based toolkit for modeling and simulating hardware-based neural networks. The proposed spiking MaxPool scheme can decide even before the complete spatiotemporal input is applied, thus selectively trading off latency with accuracy. It is observed that by allocating just 10% of the spatiotemporal input window for a pooling decision, the proposed spiking MaxPool achieves up to 61.74% accuracy with a 2-bit weight resolution in the CIFAR10 dataset classification task after training with back propagation, with only about 1% performance drop compared to 62.78% accuracy of the 100% spatiotemporal window case with the 2-bit weight resolution to reflect foundry-integrated ReRAM limitations. In addition, we propose the realization of one of the proposed spiking MaxPool techniques in an NVM crossbar array along with periphery circuits designed in a 130nm CMOS technology. The energy-efficiency estimation results show competitive performance compared to recent neuromorphic chip designs. [ABSTRACT FROM AUTHOR]

In the latest research progress, deep neural networks have been revolutionized by frameworks to extract image features more accurately. In this study, we focus on an attention model that can be useful in deep neural networks and propose a simple but strong feature extraction deep network architecture, W-Net. The architecture of our W-Net network has two mutually independent path structures, and it is designed with the following advantages. (1) There are two independent effective paths in our proposed network structure, and the two paths capture more contextual information from different scales in different ways. (2) The two paths acquire different feature images, and in the upsampling approach, we use bilinear interpolation thus reducing the feature map distortion phenomenon and integrating the different images processed. (3) The feature image processing is at a bottleneck, and a hierarchical attention module is constructed at the bottleneck by reclassifying after the channel attention module and the spatial attention module, resulting in more efficient and accurate processing of feature images. During the experiment, we also tested iSAID, a massively high spatial resolution remote sensing image dataset, with further experimental data comparison to demonstrate the generality of our method for remote sensor image segmentation. [ABSTRACT FROM AUTHOR]

The performance and serviceability of asphalt pavements have a direct influence on people's daily lives. Timely detection of pavement cracks is crucial in the task of periodic pavement survey. This paper proposes and verifies a novel computer vision-based method for recognizing pavement crack patterns. Image processing techniques, including Gaussian steerable filters, projection integrals, and image texture analyses, are employed to characterize the surface condition of asphalt pavement roads. Light Gradient Boosting Machine, Deep Neural Network, and Convolutional Neural Network are employed to recognize various patterns including longitudinal, transverse, diagonal, minor fatigue, and severe fatigue cracks. A dataset, including 12,000 samples, has been collected to construct and verify the computer vision-based approaches. Based on experiments, it can be found that all three machine learning models are capable of delivering good categorization results with an accuracy rate > 0.93 and Cohen's Kappa coefficient > 0.76. Notably, the Light Gradient Boosting Machine has achieved the most desired performance with an accuracy rate > 0.96 and Cohen's Kappa coefficient > 0.88. [ABSTRACT FROM AUTHOR]

We review the problem of automating hardware-aware architectural design process of Deep Neural Networks (DNNs). The field of Convolutional Neural Network (CNN) algorithm design has led to advancements in many fields, such as computer vision, virtual reality, and autonomous driving. The end-to-end design process of a CNN is a challenging and time-consuming task, as it requires expertise in multiple areas such as signal and image processing, neural networks, and optimization. At the same time, several hardware platforms, general-and special-purpose, have equally contributed to the training and deployment of these complex networks in a different setting. Hardware-Aware Neural Architecture Search (HW-NAS) automates the architectural design process of DNNs to alleviate human effort and generate efficient models accomplishing acceptable accuracy-performance tradeoffs. The goal of this article is to provide insights and understanding of HW-NAS techniques for various hardware platforms (MCU, CPU, GPU, ASIC, FPGA, ReRAM, DSP, and VPU), followed by the co-search methodologies of neural algorithm and hardware accelerator specifications. [ABSTRACT FROM AUTHOR]

The advancement of both the fields of Computer Vision (CV) and Artificial Neural Networks (ANNs) has enabled the development of effective automatic systems for analyzing human behavior. It is possible to recognize gestures, which are frequently used by people to communicate information non-verbally, by studying hand movements. So, the main contribution of this research is the collected dataset, which is taken from open-source videos of the relevant subjects that contain actions that depict confidence levels. The dataset contains high-quality frames with minimal bias and less noise. Secondly, we have chosen the domain of confidence determination during social issues such as interviews, discussions, or criminal investigations. Thirdly, the proposed model is a combination of two high-performing models, i.e., CNN (GoogLeNet) and LSTM. GoogLeNet is the state-of-the-art architecture for hand detection and gesture recognition. LSTM prevents the loss of information by keeping temporal data. So the combination of these two outperformed during the training and testing process. This study presents a method to recognize different categories of Self-Efficacy by performing multi-class classification based on the current situation of hand movements using visual data processing and feature extraction. The proposed architecture pre-processes the sequence of images collected from different scenarios, including humans, and their quality frames are extracted. These frames are then processed to extract and analyze the features regarding their body joints and hand position and classify them into four different classes related to efficacy, i.e., confidence, cooperation, confusion, and uncomfortable. The features are extracted using a combination framework of customized Convolutional Neural Network (CNN) layers with Long Short-Term Memory (LSTM) for feature extraction and classification. Remarkable results have been achieved from this study representing 90.48% accuracy with effective recognition of human body gestures through deep learning approaches. [ABSTRACT FROM AUTHOR]

License plate identification remains a crucial problem in computer vision, particularly in complex environments where license plates may be confused with road signs, billboards, and other objects. This paper proposes a solution by modifying the standard car–license plate–letter detection approach into a preliminary license plate detection–precise license plate detection of the four corners where the numbers are located–license plate correction–letter identification. This way, the first algorithm identifies all potential license plates and passes them as input parameters to the next algorithm for more precise detection. The main difference between this approach and other algorithms is that it uses a relatively small image compared to the whole vehicle. Thus, a small but robust network is used to find the four corners and perform a perspective transformation. This simplifies the letter recognition task for the next algorithm, as no additional transformations are required. This solution could be useful for research focusing on this specific task. It allows to apply another compact but robust neural network, increasing the overall speed of the system. Publicly available datasets were used for training and validation. The CenterNet object detection algorithm was used as a basis with a modified Hourglass-type network. The size of the network was decreased by 40% and the average accuracy was 96.19%. Speed significantly increased, reaching 2.71 ms and 405 FPS on average. [ABSTRACT FROM AUTHOR]

Plant diseases introduce significant yield and quality losses to the food production industry, worldwide. Early identification of an epidemic could lead to more effective management of the disease and potentially reduce yield loss and limit excessive input costs. Image processing and deep learning techniques have shown promising results in distinguishing healthy and infected plants at early stages. In this paper, the potential of four convolutional neural network models, including Xception, Residual Networks (ResNet)50, EfficientNetB4, and MobileNet, in the detection of rust disease on three commercially important field crops was evaluated. A dataset of 857 positive and 907 negative samples captured in the field and greenhouse environments were used. Training and testing of the algorithms were conducted using 70% and 30% of the data, respectively where the performance of different optimizers and learning rates were tested. Results indicated that EfficientNetB4 model was the most accurate model (average accuracy = 94.29%) in the disease detection followed by ResNet50 (average accuracy = 93.52%). Adaptive moment estimation (Adam) optimizer and learning rate of 0.001 outperformed all other corresponding hyperparameters. The findings from this study provide insights into the development of tools and gadgets useful in the automated detection of rust disease required for precision spraying. [ABSTRACT FROM AUTHOR]

Disasters may occur at any time and place without little to no presage in advance. With the development of surveillance and forecasting systems, it is now possible to forebode the most life-threatening and formidable disasters. However, forest fires are among the ones that are still hard to anticipate beforehand, and the technologies to detect and plot their possible courses are still in development. Unmanned Aerial Vehicle (UAV) image-based fire detection systems can be a viable solution to this problem. However, these automatic systems use advanced deep learning and image processing algorithms at their core and can be tuned to provide accurate outcomes. Therefore, this article proposed a forest fire detection method based on a Convolutional Neural Network (CNN) architecture using a new fire detection dataset. Notably, our method also uses separable convolution layers (requiring less computational resources) for immediate fire detection and typical convolution layers. Thus, making it suitable for real-time applications. Consequently, after being trained on the dataset, experimental results show that the method can identify forest fires within images with a 97.63% accuracy, 98.00% F1 Score, and 80% Kappa. Hence, if deployed in practical circumstances, this identification method can be used as an assistive tool to detect fire outbreaks, allowing the authorities to respond quickly and deploy preventive measures to minimize damage. [ABSTRACT FROM AUTHOR]

The Global Positioning System (GPS) is vulnerable to various deliberate and unintentional interferences. Therefore, identifying and coping with various interferences in this system is essential. This paper analyzes a method of reducing the dimensions of Cross Ambiguity Function (CAF) images in improving the identification of spoofing interference at the GPS using Multi-Layer Perceptron Neural Network (MLP NN) and Convolutional Neural Network (CNN). Using the proposed method reduces data complexity, which can reduce the number of learning data requirements. The simulation results indicate that, by applying the proposed image processing algorithm for different dimensions of CAF images, the CNN performs better than MLP NN in terms of training accuracy; the MLP NN is superior to CNN in terms of convergence speed of training. In addition, the results demonstrate that the operation of the proposed method is appropriate in the case of smalldelay spoofed signals. Therefore, for the intervals above 0.25 code chip, the proposed method detects spoofing attacks with a correct detection probability close to one. [ABSTRACT FROM AUTHOR]

In the past decade, deep neural networks, particularly convolutional neural networks, have revolutionised computer vision. However, all deep learning models may require a large amount of data so as to achieve satisfying results. Unfortunately, the availability of sufficient amounts of data for real-world problems is not always possible, and it is well recognised that a paucity of data easily results in overfitting. This issue may be addressed through several approaches, one of which is data augmentation. In this paper, we survey the existing data augmentation techniques in computer vision tasks, including segmentation and classification, and suggest new strategies. In particular, we introduce a way of implementing data augmentation by using local information in images. We propose a parameter-free and easy to implement strategy, the random local rotation strategy, which involves randomly selecting the location and size of circular regions in the image and rotating them with random angles. It can be used as an alternative to the traditional rotation strategy, which generally suffers from irregular image boundaries. It can also complement other techniques in data augmentation. Extensive experimental results and comparisons demonstrated that the new strategy consistently outperformed its traditional counterparts in, for example, image classification. [ABSTRACT FROM AUTHOR]

An integrated optomechanical analysis (IOA) can predict the response of an optomechanical system to temperature, gravity, vibrations, and other local loadings; thus, the normal operation of instruments under special conditions is guaranteed. Zernike polynomials are the most popular for fitting the IOA-derived mechanical deformation data. By solving the Zernike coefficients of all deformed optical surfaces, the relationship between aberrations and deformations can be further revealed. The process of IOA is summarized in this article. The principles of four primary Zernike coefficient-solving algorithms (CSAs) were expounded, and the corresponding applications are reviewed in detail, including the least squares method, the Gram–Schmidt orthogonalized method, the Householder transformation, and singular value decomposition (SVD). Artificial neural networks (ANNs) trained for solving a similar overdetermined set of equations are also discussed; an innovative Zernike CSA based on a one-dimensional convolutional neural network (1D-CNN) was proposed, emphasizing its potential for Zernike CSA. The feasibility of the neural network method was verified by conducting experiments on the primary mirror of the front reflection system of a space camera. This review can provide references for the precise optimization of IOA. [ABSTRACT FROM AUTHOR]

The creation and training of artificial neural networks with a given accuracy makes it possible to identify patterns and hidden relationships between physical and technological parameters in the production of unique building materials, predict mechanical properties, and solve the problem of detecting, classifying, and segmenting existing defects. The detection of defects of various kinds on elements of building materials at the primary stages of production can improve the quality of construction and identify the cause of particular damage. The technology for detecting cracks in building material samples is of great importance in building monitoring, in pre-venting the spread of defective material. In this paper, we consider the use of the YOLOv4 convolutional neural network for crack detection on building material samples. This was based on the creation of its own empirical database of images of samples of aerated concrete. The number of images was increased by applying our own augmentation algorithm. Optimization of the parameters of the intellectual model based on the YOLOv4 convolutional neural network was performed. Experimental results show that the YOLOv4 model developed in this article has high precision in defect detection problems: AP@50 = 85% and AP@75 = 68%. It should be noted that the model was trained on its own set of data obtained by simulating various shooting conditions, rotation angles, object deformations, and light distortions through image processing methods, which made it possible to apply the developed algorithm in practice. [ABSTRACT FROM AUTHOR]

The rapid expansion of solar industries presents unknown technological challenges. A dedicated and suitable energy forecast is an effective solution for the daily dispatching and production of the electricity grid. The traditional forecast technique uses weather and plant parameters as the model information. Nevertheless, these are insufficient to consider problematic weather variability and the various plant characteristics in the actual field. Considering the above facts and inspired by the excellent implementation of the multi-column convolutional neural network (MCNN) in image processing, we developed a novel approach for forecasting solar energy by transforming multipoint time series (MT) into images for the MCNN to examine. We first processed the data to convert the time series solar energy into image matrices. We observed that the MCNN showed a preeminent response under a ground-based high-resolution spatial–temporal image matrix with a 0.2826% and 0.5826% RMSE for 15 min-ahead forecast under clear (CR) and cloudy (CD) conditions, respectively. Our process was performed on the MATLAB deep learning platform and tested on CR and CD solar energy conditions. The excellent execution of the suggested technique was compared with state-of-the-art deep neural network solar forecasting techniques. [ABSTRACT FROM AUTHOR]

The location of electrical substations is one of the factors affecting the improvement of electrical energy distribution, as well as the management and control of this energy source. Less cost and manpower will be spent through automating the process of detection and segmentation of these features with the help of deep neural networks and the potential of existing high spatial resolution satellite images. In this study, a deep encoder-decoder neural network was used. This network is one of the most updated deep learning methods in image processing and segmentation. This network has been trained in three RGB bands with the help of high-resolution satellite images (∼1m) and eventually segmented the areas related to electrical substations with relatively high accuracy. As the results of this convolutional neural network, the IOU and Precision parameters were obtained, and their values were 88.2 and 93.7%, respectively, indicating the efficiency of the proposed deep learning method in the segmentation of existing satellite images. [ABSTRACT FROM AUTHOR]

As a low-frequency and smooth signal, the bias field has a certain destructive effect on magnetic resonance (MR) images and is the main obstacle for doctors' diagnosis and image processing (such as segmentation, texture analysis, and registration). Before analyzing a damaged MR image, a preprocessing step is required to correct the bias field in the image. Unlike traditional bias field removal algorithms based on signal models and a priori assumptions, deep learning methods do not require precisely modeling signals and bias fields and do not need to adjust parameters. An MR image with the bias field is input and the corrected MR image is output after the deep neural network being trained on a large training set. In this paper, we propose taking the original image and the local feature images of the bias field in multiple frequency bands obtained by a Log-Gabor filter bank as input, correcting the bias field of a brain MR image through a deep separable convolutional neural network. Meanwhile, to speed up the training process and improve bias correction performance, we apply residual learning and batch normalization. We conducted the same test on BrainWeb simulation database and Human Connectome Project real data set, the consistency of qualitative and quantitative evaluation shows that our training model demonstrates better performance than the traditional state-of-the-art N4 and non-iterative multi-scale (NIMS) methods. Especially for the images with high-intensity non-uniformity level, the bias field has been well corrected. [ABSTRACT FROM AUTHOR]

Medical image segmentation plays a key role in the early diagnosis and treatment of medical diseases. Thyroid nodule segmentation is a critical step in early thyroid cancer identification. Accurately segmenting thyroid nodule areas from ultrasound images is critical for clinical diagnosis and maintaining good health. Because of the fragile borders of ultrasound images and the complicated structure of thyroid tissue, it is difficult to correctly separate the delicate outlines of thyroid nodules to provide adequate segmentation findings, since they either cannot establish exact edges or segment smaller parts. The segmentation of thyroid nodule images presents some fundamental difficulties. First, the intrinsic locality of convolutional neural network models places constraints on their ability to capture information about the whole context. Second, the size of the data sets used for thyroid nodule segmentation frequently makes overfitting more likely. Finally, low-level characteristics that are important in displaying thyroid borders eventually disappear throughout the feature encoding process. We provide an effective model called Enhanced-TransUNet for thyroid nodule image segmentation to overcome these difficulties. The Transformer and UNet concepts are combined in Enhanced-TransUNet. While the UNet can successfully segregate tiny items, the Transformer can collect information about the overall environment. In order to condense superfluous characteristics and lower the chance of overfitting, Enhanced-TransUNet also makes use of an information bottleneck. Comparing our model to contemporary CNN or UNET based models, experimental findings on the TN3K and DDTI datasets for brain tumor segmentation tasks show that our model gets equivalent or better results. For the two datasets, the average Dice Score and HD95 are 82.92, 95.45, and 13.19, 1.09, respectively. Overall, the Enhanced-TransUNet model for thyroid nodule image segmentation is promising. Even with weak edges and a complicated tissue structure, it can precisely segment thyroid nodules in ultrasound pictures. Due to its usage of the information bottleneck, Enhanced-TransUNet is also less prone to overfitting than other models. As a result, an AI-based decision support system based on this model can be built to reduce workload and misdiagnosis. This system has significant potential for clinical application by radiologists and surgeons, which can increase clinical diagnostic accuracy and efficiency. • A novel segmentation model called Enhanced-TransUNet to separate thyroid nodules in ultrasound. • Integrate vision transformers with UNet. • Enhanced-TransUNet shows promising segmentation accuracy on image segmentation. • We present its efficacy through both qualitative analysis and numerical measures. [ABSTRACT FROM AUTHOR]

Charts are often used for the graphical representation of tabular data. Due to their vast expansion in various fields, it is necessary to develop computer algorithms that can easily retrieve and process information from chart images in a helpful way. Convolutional neural networks (CNNs) have succeeded in various image processing and classification tasks. Nevertheless, the success of training neural networks in terms of result accuracy and computational requirements requires careful construction of the network layers' and networks' parameters. We propose a novel Shallow Convolutional Neural Network (SCNN) architecture for chart-type classification and image generation. We validate the proposed novel network by using it in three different models. The first use case is a traditional SCNN classifier where the model achieves average classification accuracy of 97.14%. The second use case consists of two previously introduced SCNN-based models in parallel, with the same configuration, shared weights, and parameters mirrored and updated in both models. The model achieves average classification accuracy of 100%. The third proposed use case consists of two distinct models, a generator and a discriminator, which are both trained simultaneously using an adversarial process. The generated chart images are plausible to the originals. Extensive experimental analysis end evaluation is provided for the classification task of seven chart classes. The results show that the proposed SCNN is a powerful tool for chart image classification and generation, comparable with Deep Convolutional Neural Networks (DCNNs) but with higher efficiency, reduced computational time, and space complexity. [ABSTRACT FROM AUTHOR]

Objective: We aimed to develop a computer-aided detection (CAD) system for accurate identification of benign pigmented skin lesions (PSLs) from images captured using a digital camera or a smart phone. Methods: We collected a total of 12,836 clinical images which had been classified and location-labeled for training and validating. Four models were developed and validated; you only look once, v4 (YOLOv4), you only look once, v5 (YOLOv5), single shot multibox detector (SSD) and faster region-based convolutional neural networks (Faster R-CNN). The performance of the models was compared with three trained dermatologists, respectively. The accuracy of the best model was further tested and validated using smartphone-captured images. Results: The accuracies of YOLOv4, YOLOv5, SSD and Faster R-CNN were 0.891, 0.929, 0.852 and 0.874, respectively. The precision, sensitivity and specificity of YOLOv5 (the best model) were 0.956, 0.962 and 0.952, respectively. The accuracy of YOLOv5 model for images captured using a smart-phone was 0.905. The CAD based YOLOv5 system can potentially be used in clinical identification of PSLs. Conclusion: We developed and validated a CAD system for automatic identification of benign PSLs using digital images. This approach may be used by non-dermatologists to easily diagnose by taking a photo of skin lesion and guide on management of PSLs. [ABSTRACT FROM AUTHOR]

Object detection is an essential task in computer vision and image processing. It has many applications in various domains like medical diagnosis, civil military, video surveillance, security, etc. In some vision-related approaches, object detection used as an integral part, such as semantic segmentation, instance segmentation, pose estimation, suspicious activity detection, etc. The first stage in the pipeline is to detect an object. The survey begins with significant highlights of deep learning for object detection. It provides a comprehensive study on object representation; Convolution Neural Network (CNN) and different Deep Convolution Neural Network architecture. It presents a concise review of renowned datasets and definitive measurement metrics, forming the primitive baseline to evaluate the detection framework. Then studies in detail on detection framework one-stage and two-stage detectors and evaluates each framework with standard datasets listing its vital significance. The study also explores different issues of object detection like multi-scale, intra-class variations, generalization & security. Moreover, lists the primitive steps for creating object detector for different conditions from the reviewed survey. Finally, it proposes promising research directions. [ABSTRACT FROM AUTHOR]

Chest and lung diseases are among the most serious chronic diseases in the world, and they occur as a result of factors such as smoking, air pollution, or bacterial infection, which would expose the respiratory system and chest to serious disorders. Chest diseases lead to a natural weakness in the respiratory system, which requires the patient to take care and attention to alleviate this problem. Countries are interested in encouraging medical research and monitoring the spread of communicable diseases. Therefore, they advised researchers to perform studies to curb the diseases' spread and urged researchers to devise methods for swiftly and readily detecting and distinguishing lung diseases. In this paper, we propose a hybrid architecture of contrast-limited adaptive histogram equalization (CLAHE) and deep convolutional network for the classification of lung diseases. We used X-ray images to create a convolutional neural network (CNN) for early identification and categorization of lung diseases. Initially, the proposed method implemented the support vector machine to classify the images with and without using CLAHE equalizer. The obtained results were compared with the CNN networks. Later, two different experiments were implemented with hybrid architecture of deep CNN networks and CLAHE as a preprocessing for image enhancement. The experimental results indicate that the suggested hybrid architecture outperforms traditional methods by roughly 20% in terms of accuracy. [ABSTRACT FROM AUTHOR]

This paper contains the specification of functional and nonfunctional requirements of a CNN (Convolutional Neural Network) model to identify masses in liver ultrasounds. A neural network model based on CNN was implemented, which performs the analysis of ultrasound images of the liver, identifying the presence of masses or not in them. CNN can be executed either on a machine with a CPU (Central Processing Unit) or in a high processing capacity environment that uses GPU (Graphics Processing Unit). The validation of the accuracy of the obtained results happens through the statistical technique of k-fold cross-validation using IoT. During the neural network training process, dense data necessitates a lot of computational resources. Because the number of network entries is precisely proportional to the number of pixels in the image, parallel processing for image processing is justified. It is important to remember that medical photos cannot be saved in a lossy format and must be identified with exceptional accuracy using IoT. [ABSTRACT FROM AUTHOR]

Summary: Fluorescence microscopy is common in bacteria–plant interaction studies. However, strong autofluorescence from plant tissues impedes in vivo studies on endophytes tagged with fluorescent proteins. To solve this problem, we developed a deep‐learning‐based approach to eliminate plant autofluorescence from fluorescence microscopy images, tested for the model endophyte Azoarcus olearius BH72 colonizing Oryza sativa roots.Micrographs from three channels (tdTomato for gene expression, green fluorescent protein (GFP) and AutoFluorescence (AF)) were processed by a neural network based approach, generating images that simulate the background autofluorescence in the tdTomato channel. After subtracting the model‐generated signals from each pixel in the genuine channel, the autofluorescence in the tdTomato channel was greatly reduced or even removed.The deep‐learning‐based approach can be applied for fluorescence detection and quantification, exemplified by a weakly expressed, a cell‐density modulated and a nitrogen‐fixation gene in A. olearius. A transcriptional nifH::tdTomato fusion demonstrated stronger induction of nif genes inside roots than outside, suggesting extension of the rhizosphere effect for diazotrophs into the endorhizosphere.The pre‐trained convolutional neural network model is easily applied to process other images of the same plant tissues with the same settings. This study showed the high potential of deep‐learning‐based approaches in image processing. With proper training data and strategies, autofluorescence in other tissues or materials can be removed for broad applications. [ABSTRACT FROM AUTHOR]

Mud layer height of thickener is the key quality index of thickening process which is difficult to achieve real-time detection with existing methods in reality. While the need of developing a soft sensor model which can be used for real-time detection of mud layer height, we proposed an end-to-end mud layer height prediction method with attention mechanism-based convolutional neural network (CNN). The dynamic features are firstly extracted from the image samples based on CNN, and then two types of attention mechanism are embedded sequentially to contribute to more precise prediction results. Compared with the traditional spatial attention mechanism, the regional spatial attention mechanism we proposed selectively divides the spatial feature map into regions, while regions containing important features are assigned larger weights. Adding the channel and regional spatial attention mechanism in CNN not only effectively improve both the precision and calculation speed, but also affect the dimension of the output feature map, so as to avoid the loss of channel or spatial attention information of the feature map. To verify the validity of the proposed method, different attention mechanisms are embedded in the CNN, and the corresponding experiments are carried out on the dataset of the thickener mud layer. The experimental results demonstrate the feasibility and effectiveness of the mud layer height prediction method. • A novel convolution neural network model combined with attention mechanism is designed for mud layer height of the thickener prediction. • A regional spatial attention mechanism is proposed which divides the feature map into regions in the spatial dimension according to the distribution of features, and the weights of each region are constructed to increase the weights of the regions containing the key information. • Applying the proposed method of DAM with region divided to the thickening process improves the network features extraction capability, and the prediction accuracy. [ABSTRACT FROM AUTHOR]

The novel coronavirus (COVID-19) is a contagious viral disease that has rapidly spread worldwide since December 2019, causing the disruption of life and heavy economic losses. Since the beginning of the virus outbreak, a polymerase chain reaction has been used to detect the virus. However, since it is an expensive and slow method, artificial intelligence researchers have attempted to develop quick, inexpensive alternative methods of diagnosis to help doctors identify positive cases. Therefore, researchers are starting to incorporate chest X-ray scans (CXRs), an easy and inexpensive examination method. This study used an approach that uses image cropping methods and a deep learning technique (updated VGG16 model) to classify three public datasets. This study had four main steps. First, the data were split into training and testing sets (70% and 30%, respectively). Second, in the image processing step, each image was cropped to show only the chest area. The images were then resized to 150 × 150. The third step was to build an updated VGG16 convolutional neural network (VGG16-CNN) model using multiple classifications (three classes: COVID-19, normal, and pneumonia) and binary classification (COVID-19 and normal). The fourth step was to evaluate the model's performance using accuracy, sensitivity, and specificity. This study obtained 97.50% accuracy for multiple classifications and 99.76% for binary classification. The study also got the best COVID-19 classification accuracy (99%) for both models. It can be considered that the scientific contribution of this research is summarized as: the VGG16 model was reduced from approximately 138 million parameters to around 40 million parameters. Further, it was tested on three different datasets and proved highly efficient in performance. [ABSTRACT FROM AUTHOR]

Crack detection plays a pivotal role in structural health monitoring. Deep convolutional neural networks (DCNN) provide a way to achieve image classification efficiently and accurately due to their powerful image processing ability. In this paper, we propose a semi-supervised learning method based on a DCNN to achieve anomaly crack detection. In the proposed method, the training set for the network only requires a small number of normal (non-crack) images but can achieve high detection accuracy. Moreover, the trained model has strong robustness in the condition of uneven illumination and evident crack difference. The proposed method is applied to the images of walls, bridges and pavements, and the results show that the detection accuracy comes up to 99.48%, 92.31% and 97.57%, respectively. In addition, the features of the neural network can be visualized to describe its working principle. This method has great potential in practical engineering applications. [ABSTRACT FROM AUTHOR]

Partial shading conditions (PSC) have negative effects on the operation of photovoltaic (PV) systems. In this paper, a PV array reconfiguration method is developed to minimize power losses of PV arrays under partial shading conditions. The proposed reconfiguration method is based on equalizing the reduction of the short-circuit current of the PV modules in the PV array. Eight state-of-the-art Convolutional Neural Network models are employed to estimate the effect of shading on the short-circuit current of a PV module. These models include LeNet-5, AlexNet, VGG 11, VGG 19, Inception V3, ResNet 18, ResNet 34, and ResNet 50. Among eight models, the VGG 19 achieves the best accuracy on 1842 sample images. Therefore, this model is used to estimate the ratio of the actual short-circuit current and the estimated short-circuit current in four studied shading scenarios. This ratio decides the switching rule between PV modules throughout the PV array under PSC. A 2 × 2 experimental PV array shows that the proposed reconfiguration method improves the output power from 5.81% to 25.19% in four shading patterns. Accordingly, the power losses are reduced from 1.32% to 13.75%. The power improvement and the reduction of power losses of the proposed dynamic PV array reconfiguration system under four case studies demonstrates its effectiveness in addressing the effects of PSC on the PV array. [ABSTRACT FROM AUTHOR]

Coronavirus disease 2019 (COVID-19) has a significant impact on human life. The novel pandemic forced humans to change their lifestyles. Scientists have broken through the vaccine in many countries, but the face mask is the only protection for public interaction. In this study, deep neural networks (DNN) have been employed to determine the persons wearing masks correctly. The faster region-based convolutional neural networks (RCNN) model has been used to train the data using graphics processing unit (GPU) device. To achieve our goals, we used a multiphase detection model: first, to label the face mask, and second to detect the edge and compute edge projection for the chosen face region within the face mask. The current findings revealed that faster RCNN was efficient and precise, giving 97% accuracy. The overall loss after 200,000 epochs is 0.0503, with a trend to decrease. While the loss is falling, we are getting more accurate results. As a result, the faster RCNN technique effectively identifies whether a person is wearing face masks or not, and the training period was decreased with better accuracy. In the future, Deep Neural Network (DNN) might be used first to train the data and then compress the dimensions of the input to run it on low-powered devices, resulting in a lower computational cost. Our proposed system can achieve high face detection accuracy and coarsely obtain face posture estimation based on the specified rule. The faster RCNN learning algorithm returns high precision, and the model's lower computational cost is achieved on GPU. We use the "label-image" application to label the photographs extracted from the dataset and apply Inception V2 of faster RCNN for face mask detection and classification. [ABSTRACT FROM AUTHOR]

With the growing demand for geospatial data, challenging aerial images with high spatial, spectral, and temporal resolution achieve excellent development. Currently, deep Convolutional Neural Network (CNN) structures are applied widely for object detection. Nevertheless, existing deep CNN-based models consist of complex network structures and require immense amounts of graphics processing unit (GPU) computation power with high energy consumption. Thus, achieving efficient real-time object detection for limited memory and processing capacity embedded device is a major challenge. This paper proposes a feasible and lightweight object detection model based on deep CNN where a mobile inverted bottleneck module is adopted in the backbone structure. Moreover, an enhanced spatial pyramid pooling is adopted to increase the receptive field in the network by concatenating the multi-scale local region features. The experimental results demonstrated that the proposed model achieved higher average precision and required the smallest memory storage compared to previous works. Moreover, the proposed model offers the best trade-offs in terms of detection accuracy, model size, and detection time which has excellent potential to be deployed on limited capacity embedded device. [ABSTRACT FROM AUTHOR]