Neural Networks In Computer Vision Research Articles

Fractional Calculus Meets Neural Networks for Computer Vision: A Survey

Traditional computer vision techniques aim to extract meaningful information from images but often depend on manual feature engineering, making it difficult to handle complex real-world scenarios. Fractional calculus (FC), which extends derivatives to non-integer orders, provides a flexible way to model systems with memory effects and long-term dependencies, making it a powerful tool for capturing fractional rates of variation. Recently, neural networks (NNs) have demonstrated remarkable capabilities in learning complex patterns directly from raw data, automating computer vision tasks and enhancing performance. Therefore, the use of fractional calculus in neural network-based computer vision is a powerful method to address existing challenges by effectively capturing complex spatial and temporal relationships in images and videos. This paper presents a survey of fractional calculus neural network-based (FC NN-based) computer vision techniques for denoising, enhancement, object detection, segmentation, restoration, and NN compression. This survey compiles existing FFC NN-based approaches, elucidates underlying concepts, and identifies open questions and research directions. By leveraging FC’s properties, FC NN-based approaches offer a novel way to improve the robustness and efficiency of computer vision systems.

Understanding Wireless ChannelsThrough NeRF2

Despite Maxwell's formulation of the electromagnetic wave laws over a century and a half ago, accurately modeling the transmission of RF signals within electrically complex environments continues to be a formidable challenge. This complexity arises from intricate interactions of RF signals with various obstacles, such as reflections and diffractions. Inspired by the effective utilization of neural networks in computer vision for representing optical fields, this paper introduces NeRF2, a novel neural radio-frequency radiance field. NeRF2 conceptualizes a continuous, volumetric scene function to interpret RF signal propagation. Uniquely, NeRF2, once trained with limited signal measurements, can predict the characteristics and reception of a signal at an arbitrary point, given the transmitter's location. Functioning as a physical-layer neural network, NeRF2 amalgamates learned statistical models with physical ray tracing to create synthetic datasets, tailored for the training needs of application-layer artificial neural networks (ANNs). This integration, termed "turbo-learning," combines authentic and synthetic datasets, significantly augmenting ANNs' training efficacy. Experimental results illustrate that turbo-learning can substantially improve performance, with an estimated 50% enhancement. The efficacy of NeRF2 is further underscored in its applications in indoor localization and 5G MIMO systems.

A review of convolutional neural networks in computer vision

In computer vision, a series of exemplary advances have been made in several areas involving image classification, semantic segmentation, object detection, and image super-resolution reconstruction with the rapid development of deep convolutional neural network (CNN). The CNN has superior features for autonomous learning and expression, and feature extraction from original input data can be realized by means of training CNN models that match practical applications. Due to the rapid progress in deep learning technology, the structure of CNN is becoming more and more complex and diverse. Consequently, it gradually replaces the traditional machine learning methods. This paper presents an elementary understanding of CNN components and their functions, including input layers, convolution layers, pooling layers, activation functions, batch normalization, dropout, fully connected layers, and output layers. On this basis, this paper gives a comprehensive overview of the past and current research status of the applications of CNN models in computer vision fields, e.g., image classification, object detection, and video prediction. In addition, we summarize the challenges and solutions of the deep CNN, and future research directions are also discussed.

Tourism image classification based on convolutional neural network SqueezeNet--Taking Slender West Lake as an example.

Tourism image classification plays an important role in the study of clarifying the real perception of tourism resources by tourists, which cannot be studied in depth by human vision alone. The development of convolutional neural networks in computer vision brings new opportunities for tourism image classification research. In this study, SqueezeNet, a lightweight convolutional neural network, was selected and improved on the basis of the original model for 3740 Slender West Lake tourism image datasets. It is found that the validation accuracy of the model is up to 85.75%, and the size is only 2.64 MB, which is a good classification effect. This reduces the parameters while ensuring high accuracy classification of tourism images, providing a more scientific reference for the study of tourism images and pointing out a new direction for the development and planning of tourism resources.

Artificial intelligence and machine learning applications in the Spanish nuclear field

Machine Learning and Artificial Intelligence techniques are increasingly applied in the nuclear field. In this way, Spanish companies and research institutions have been developing different projects in the field of nuclear energy applying such techniques. A brief overview of some of these projects is presented, which cover areas as diverse as nuclear data, nuclear fuel manufacturing, simulation codes, fuel performance monitoring, nuclear inspection, and operation training. Among the techniques used to develop them are deep neural networks for building surrogate models, convolutional neural networks for computer vision, Bayesian networks and Gaussian processes for predictive models, and speech recognition with large language models.

Renyi entropy analysis of a deep convolutional representation for texture recognition

Despite the recent success of convolutional neural networks in computer vision in general, texture images still pose an important challenge to those models, especially when dealing with textures “in the wild”. In this context, deep learning models can benefit from the introduction of features long known to be useful for texture modeling. And that is the case of entropy, a measure of texture regularity that has played an important role in classical computer vision. Based on this observation, here we propose an alternative analysis over deep convolutional neural features based on entropy for texture representation and, particularly, texture classification. More precisely, we couple a module to the convolutional backbone that locally computes the Renyi entropy of the latent representation at multiple levels. The rationale for using Renyi entropy is essentially two-fold: (1) It connects the concept of entropy with multifractal theory, another well explored measure especially in domains of application where we seek some physical interpretation of the descriptors, e.g., in medicine, biology, and others; (2) It has an extra degree of freedom (α parameter) that can be fine-tuned. The main contribution of this study is the development of a strategy that can improve the performance of convolutional neural networks in texture recognition tasks, adding low computational cost. The effectiveness of our method is verified in texture classification of benchmark datasets, as well as in a practical task of plant species identification. Our method achieves a competitive accuracy of 84.5% in the classification of KTH-TIPS-2b and 80.9% in FMD, which are two challenging benchmark databases. Besides, an accuracy of 91.6% is obtained in the plant identification application, to our knowledge outperforming all the results previously published on this task. In both scenarios, the proposed descriptors outperform several approaches from the state-of-the-art, confirming the method potential as a rich and robust solution for texture analysis in general. The results also suggest that information-theoretical measures like entropy can be a reliable source of information to compose a precise and robust latent representation of texture images.

Power Supply Technologies for Drones and Machine Vision Applications: A Comparative Analysis and Future Trends

The field of Unmanned Aerial Vehicles (UAVs), or drones, is encountering quick development in the areas of air transportation and computerization. Progress in innovation has prompted more noteworthy capacities and highlights in UAVs, which are currently broadly involved by the military and flying industry for an assortment of high-end generally safe errands. Highly advanced UAVs that can be controlled remotely via a controller, mobile phone, or ground station cockpit have been developed through the integration of automation technology and machine vision, which includes thermal imaging, cameras, sensors, and other sensors. The three primary characteristics of UAVs will be investigated in this study, namely power-source technology, deep-learning neural networks for computer vision, and some of the applications that are used the most. The goal is to thoroughly examine these characteristics and offer suggestions for addressing some of the difficulties of optimizing UAV performance and also exploring potential future trends.

A Presentation of Structures and Applications of Convolutional Neural Networks

This paper reviewed the history of convolutional neural networks, why and how they developed, and what inspired the scientists to design them. To make CNNs simpler to understand, we wrote about their characteristics and structures while introducing the basic units of convolutional neural networks, including training and modeling parameters and how they would affect the confidence and efficiency of the whole process, different kinds of layers and how they work, multiple pooling methods and loss functions with their formulas. This paper also included applications of convolutional neural networks in computer vision and natural language processing while specifying and analyzing the technologies in use to clarify this introduction. Challenges and future research directions of the convolutional neural networks were pointed out to help refine this technique.

Adversarial Machine Learning for Network Intrusion Detection Systems: A Comprehensive Survey

Network-based Intrusion Detection System (NIDS) forms the frontline defence against network attacks that compromise the security of the data, systems, and networks. In recent years, Deep Neural Networks (DNNs) have been increasingly used in NIDS to detect malicious traffic due to their high detection accuracy. However, DNNs are vulnerable to adversarial attacks that modify an input example with imperceivable perturbation, which causes a misclassification by the DNN. In security-sensitive domains, such as NIDS, adversarial attacks pose a severe threat to network security. However, existing studies in adversarial learning against NIDS directly implement adversarial attacks designed for Computer Vision (CV) tasks, ignoring the fundamental differences in the detection pipeline and feature spaces between CV and NIDS. It remains a major research challenge to launch and detect adversarial attacks against NIDS. This article surveys the recent literature on NIDS, adversarial attacks, and network defences since 2015 to examine the differences in adversarial learning against deep neural networks in CV and NIDS. It provides the reader with a thorough understanding of DL-based NIDS, adversarial attacks and defences, and research trends in this field. We first present a taxonomy of DL-based NIDS and discuss the impact of taxonomy on adversarial learning. Next, we review existing white-box and black-box adversarial attacks on DNNs and their applicability in the NIDS domain. Finally, we review existing defence mechanisms against adversarial examples and their characteristics.

Rearranging Pixels is a Powerful Black-Box Attack for RGB and Infrared Deep Learning Models

Recent research has found that neural networks for computer vision are vulnerable to several types of external attacks that modify the input of the model, with the malicious intent of producing a misclassification. With the increase in the number of feasible attacks, many defence approaches have been proposed to mitigate the effect of these attacks and protect the models. Mainly, the research on both attack and defence has focused on RGB images, while other domains, such as the infrared domain, are currently underexplored. In this paper, we propose two attacks, and we evaluate them on multiple datasets and neural network models, showing that the results outperform others established attacks, on both RGB as well as infrared domains. In addition, we show that our proposal can be used in an adversarial training protocol to produce more robust models, with respect to both adversarial attacks and natural perturbations that can be applied to input images. Lastly, we study if a successful attack in a domain can be transferred to an aligned image in another domain, without any further tuning. The code, containing all the files and the configurations used to run the experiments, is available <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/jaryP/IR-RGB-domain-attackonline</uri> .

Black-box error diagnosis in Deep Neural Networks for computer vision: a survey of tools

The application of Deep Neural Networks (DNNs) to a broad variety of tasks demands methods for coping with the complex and opaque nature of these architectures. When a gold standard is available, performance assessment treats the DNN as a black box and computes standard metrics based on the comparison of the predictions with the ground truth. A deeper understanding of performances requires going beyond such evaluation metrics to diagnose the model behavior and the prediction errors. This goal can be pursued in two complementary ways. On one side, model interpretation techniques "open the box" and assess the relationship between the input, the inner layers and the output, so as to identify the architecture modules most likely to cause the performance loss. On the other hand, black-box error diagnosis techniques study the correlation between the model response and some properties of the input not used for training, so as to identify the features of the inputs that make the model fail. Both approaches give hints on how to improve the architecture and/or the training process. This paper focuses on the application of DNNs to Computer Vision (CV) tasks and presents a survey of the tools that support the black-box performance diagnosis paradigm. It illustrates the features and gaps of the current proposals, discusses the relevant research directions and provides a brief overview of the diagnosis tools in sectors other than CV.

Convolutional Neural Networks in Computer Vision for Grain Crop Phenotyping: A Review

Computer vision (CV) combined with a deep convolutional neural network (CNN) has emerged as a reliable analytical method to effectively characterize and quantify high-throughput phenotyping of different grain crops, including rice, wheat, corn, and soybean. In addition to the ability to rapidly obtain information on plant organs and abiotic stresses, and the ability to segment crops from weeds, such techniques have been used to detect pests and plant diseases and to identify grain varieties. The development of corresponding imaging systems to assess the phenotypic parameters, yield, and quality of crop plants will increase the confidence of stakeholders in grain crop cultivation, thereby bringing technical and economic benefits to advanced agriculture. Therefore, this paper provides a comprehensive review of CNNs in computer vision for grain crop phenotyping. It is meaningful to provide a review as a roadmap for future research in such a thriving research area. The CNN models (e.g., VGG, YOLO, and Faster R-CNN) used CV tasks including image classification, object detection, semantic segmentation, and instance segmentation, and the main results of recent studies on crop phenotype detection are discussed and summarized. Additionally, the challenges and future trends of the phenotyping techniques in grain crops are presented.

Incremental learning with neural networks for computer vision: a survey

Incremental learning with neural networks for computer vision: a survey

Smart Gravimetric System for Enhanced Security of Accesses to Public Places Embedding a MobileNet Neural Network Classifier

In this article, we present a smart gravimetric system for the automatic security monitoring of the accesses to public places with some entrance ticket or pass required (e.g., railway stations, subway stations, museums, exhibitions). The main objective is to spot illegal behaviors, for example, two people trying to enter using only one ticket; simultaneously, other events of interest can also be detected, for example, the passage of a person with disabilities or a person with a stroller. These pieces of information can be used to arrange the appropriate assistance or to conduct statistical analyses on the monitored place frequentation. The proposed system is based on a robust gravimetric footboard, composed of steel plates laid on load cells. The data processing, classification included, is totally executed on embedded STM32 microcontrollers, following the edge-computing paradigm. The classification exploits a MobileNetV2, a class of neural networks for computer vision, especially created for embedded and mobile applications. We realized a prototype of the gravimetric system for laboratory testing and we acquired a set of samples for the considered event classes to assess the MobileNet network capabilities. We compared the performance of this implementation with a more classic VGG16 convolutional neural network (CNN) running on a PC machine. We show that in the face of akin classification results and similar execution times between the two, the embedded MobileNet version brings many advantages, namely portability, cost, size, energy consumption reduction, and finally simplification in the installation procedure.

Analysis of Explainers of Black Box Deep Neural Networks for Computer Vision: A Survey

Deep Learning is a state-of-the-art technique to make inference on extensive or complex data. As a black box model due to their multilayer nonlinear structure, Deep Neural Networks are often criticized as being non-transparent and their predictions not traceable by humans. Furthermore, the models learn from artificially generated datasets, which often do not reflect reality. By basing decision-making algorithms on Deep Neural Networks, prejudice and unfairness may be promoted unknowingly due to a lack of transparency. Hence, several so-called explanators, or explainers, have been developed. Explainers try to give insight into the inner structure of machine learning black boxes by analyzing the connection between the input and output. In this survey, we present the mechanisms and properties of explaining systems for Deep Neural Networks for Computer Vision tasks. We give a comprehensive overview about the taxonomy of related studies and compare several survey papers that deal with explainability in general. We work out the drawbacks and gaps and summarize further research ideas.

REMOTE SENSING IMAGE CLASSIFICATION WITH THE SEN12MS DATASET

Abstract. Image classification is one of the main drivers of the rapid developments in deep learning with convolutional neural networks for computer vision. So is the analogous task of scene classification in remote sensing. However, in contrast to the computer vision community that has long been using well-established, large-scale standard datasets to train and benchmark high-capacity models, the remote sensing community still largely relies on relatively small and often application-dependend datasets, thus lacking comparability. With this paper, we present a classification-oriented conversion of the SEN12MS dataset. Using that, we provide results for several baseline models based on two standard CNN architectures and different input data configurations. Our results support the benchmarking of remote sensing image classification and provide insights to the benefit of multi-spectral data and multi-sensor data fusion over conventional RGB imagery.

ResTS: Residual Deep interpretable architecture for plant disease detection

Recently many methods have been induced for plant disease detection by the influence of Deep Neural Networks in Computer Vision. However, the dearth of transparency in these types of research makes their acquisition in the real-world scenario less approving. We propose an architecture named ResTS (Residual Teacher/Student) that can be used as visualization and a classification technique for diagnosis of the plant disease. ResTS is a tertiary adaptation of formerly suggested Teacher/Student architecture. ResTS is grounded on a Convolutional Neural Network (CNN) structure that comprises two classifiers (ResTeacher and ResStudent) and a decoder. This architecture trains both the classifiers in a reciprocal mode and the conveyed representation between ResTeacher and ResStudent is used as a proxy to envision the dominant areas in the image for categorization. The experiments have shown that the proposed structure ResTS (F1 score: 0.991) has surpassed the Teacher/Student architecture (F1 score: 0.972) and can yield finer visualizations of symptoms of the disease. Novel ResTS architecture incorporates the residual connections in all the constituents and it executes batch normalization after each convolution operation which is dissimilar to the formerly proposed Teacher/Student architecture for plant disease diagnosis. Residual connections in ResTS help in preserving the gradients and circumvent the problem of vanishing or exploding gradients. In addition, batch normalization after each convolution operation aids in swift convergence and increased reliability. All test results are attained on the PlantVillage dataset comprising 54 306 images of 14 crop species.

An Acquisition Parameter Study for Machine-Learning-Enabled Electron Backscatter Diffraction.

Methods within the domain of artificial intelligence are gaining traction for solving a range of materials science objectives, notably the use of deep neural networks for computer vision for the analysis of electron diffraction patterns. An important component of deploying these models is an understanding of the performance as experimental diffraction conditions are varied. This knowledge can inspire confidence in the classifications over a range of operating conditions and identify where performance is degraded. Elucidating the relative impact of each parameter will suggest the most important parameters to vary during the collection of future training data. Knowing which data collection efforts to prioritize is of concern given the time required to collect or simulate vast libraries of diffraction patterns for a wide variety of materials without considering varying any parameters. In this work, five parameters, frame averaging, detector tilt, sample-to-detector distance, accelerating voltage, and pattern resolution, essential to electron diffraction are individually varied during the collection of electron backscatter diffraction patterns to explore the effect on the classifications produced by a deep neural network trained from diffraction patterns captured using a fixed set of parameters. The model is shown to be resilient to nearly all the individual changes examined here.

IX-ResNet: fragmented multi-scale feature fusion for image classification

With the continuous in-depth study of convolutional neural network in computer vision, how to improve the performance of network structure has been the focus of current research. Recent works have shown that multi-scale feature concatenation, shortcut connection and grouping convolution can effectively train deeper networks and improve the accuracy and effectiveness of the network. In this paper, we present a novel feature transformation strategy of fragmented multi-scale feature fusion. Moreover, an efficient modularized image classification network, IX-ResNet, is proposed based on this new strategy. IX-ResNet consists of many large isomorphic modules stacked in the form of residual network while Each large module can be composed of many small heterogeneous modules. The performance of IX-ResNet is verified on cifar-10, cifar-100 and ImageNet-1 K datasets, which indicates that IX-ResNet model using fragmented multi-scale feature fusion strategy can further improve accuracy compare to the original grouping convolution network ResNeXt with the same or even lower parameters.

Novel Convergence Results of Adaptive Stochastic Gradient Descents.

Adaptive stochastic gradient descent, which uses unbiased samples of the gradient with stepsizes chosen from the historical information, has been widely used to train neural networks for computer vision and pattern recognition tasks. This paper revisits the theoretical aspects of two classes of adaptive stochastic gradient descent methods, which contain several existing state-of-the-art schemes. We focus on the presentation of novel findings: In the general smooth case, the nonergodic convergence results are given, that is, the expectation of the gradients' norm rather than the minimum of past iterates is proved to converge; We also studied their performances under Polyak-Łojasiewicz property on the objective function. In this case, the nonergodic convergence rates are given for the expectation of the function values. Our findings show that more substantial restrictions on the steps are needed to guarantee the nonergodic function values' convergence (rates).