A subset of machinelearning algorithm called Deep Reinforcement learning (DRL) enables computers or agents to learn behavior by taking actions in a given environment through trial and error while observing the reward...
详细信息
A subset of machinelearning algorithm called Deep Reinforcement learning (DRL) enables computers or agents to learn behavior by taking actions in a given environment through trial and error while observing the rewards. In this learning paradigm, the agent is given a set of actions to chose and is then rewarded or punished depending on the results of those actions. The agent gradually develops the ability to make the best decisions by maximizing its rewards. DRL blends the learning ability of deep neural networks into the decision making capability of reinforcement learning (RL) frameworks in order to seeks and identify the most favorable set of actions. This survey paper studies DRL applications for diverse imageprocessing tasks. It starts by providing an overview of the latest model-free and model-based RL and DRL algorithms. Then, it looks at how DRL is being used for various imageprocessing tasks including image segmentation and classification, object detection, image registration, image denoising, image restoration, and landmark detection. Lastly, the paper discusses the potential uses and challenges of DRL in the proposed area by addressing the research questions. Survey results have showed that DRL is a promising approach for imageprocessing and that it has the potential to solve complex imageprocessing tasks.
This paper presents the design of a comprehensive automatic fish processing line utilizing machinelearningalgorithms. The processing line encompasses several essential steps, including fish identification by type, f...
详细信息
This paper presents the design of a comprehensive automatic fish processing line utilizing machinelearningalgorithms. The processing line encompasses several essential steps, including fish identification by type, fish sorting by size, fish orientation based on shape, and fish cutting at the optimal chopping points. The primary objective of this design is not just automation but also maximizing economic benefits by preserving the maximum amount of fish meat during the cutting process, achieved through the application of machinelearningalgorithms. To accomplish these goals, we employ a combination of transfer learning and convolutional neural networks to establish criteria for actions across all stages of automatic fish processing. At the heart of the processing station is a conveyor belt equipped with numerous sensors and lenses. Positioned along this conveyor belt are two robotic arms, responsible for precise positioning and cutting operations, all guided by the machinelearningalgorithms. To provide a visual representation of these design concepts, we have created a 3D SolidWorks model.
As the 16th most common cancer globally, oral cancer yearly accounts for some 355,000 new cases. This study underlines that an early diagnosis can improve the prognosis and cut down on mortality. It discloses a multif...
详细信息
As the 16th most common cancer globally, oral cancer yearly accounts for some 355,000 new cases. This study underlines that an early diagnosis can improve the prognosis and cut down on mortality. It discloses a multifaceted approach to the detection of oral cancer, including clinical examination, biopsies, imaging techniques, and the incorporation of artificial intelligence and deep learning methods. This study is distinctive in that it provides a thorough analysis of the most recent AI-based methods for detecting oral cancer, including deep learning models and machine learningalgorithms that use convolutional neural networks. By improving the precision and effectiveness of cancer cell detection, these models eventually make early diagnosis and therapy possible. This study also discusses the importance of techniques in image pre-processing and segmentation in improving image quality and feature extraction, an essential component of accurate diagnosis. These techniques have shown promising results, with classification accuracies reaching up to 97.66% in some models. Integrating the conventional methods with the cutting-edge AI technologies, this study seeks to advance early diagnosis of oral cancer, thus enhancing patient outcomes and cutting down on the burden this disease is imposing on healthcare systems.
The increase in precision agriculture has promoted the development of picking robot technology,and the visual recognition system at its core is crucial for improving the level of agricultural *** paper reviews the pro...
详细信息
The increase in precision agriculture has promoted the development of picking robot technology,and the visual recognition system at its core is crucial for improving the level of agricultural *** paper reviews the progress of visual recognition tech-nology for picking robots,including image capture technology,target detection algorithms,spatial positioning strategies and scene *** article begins with a description of the basic structure and function of the vision system of the picking robot and em-phasizes the importance of achieving high-efficiency and high-accuracy recognition in the natural agricultural ***-sequently,various imageprocessing techniques and vision algorithms,including color image analysis,three-dimensional depth percep-tion,and automatic object recognition technology that integrates machinelearning and deep learningalgorithms,were *** the same time,the paper also highlights the challenges of existing technologies in dynamic lighting,occlusion problems,fruit maturity di-versity,and real-time processing *** paper further discusses multisensor information fusion technology and discusses methods for combining visual recognition with a robot control system to improve the accuracy and working rate of *** the same time,this paper also introduces innovative research,such as the application of convolutional neural networks(CNNs)for accurate fruit detection and the development of event-based vision systems to improve the response speed of the *** the end of this paper,the future development of visual recognition technology for picking robots is predicted,and new research trends are proposed,including the refinement of algorithms,hardware innovation,and the adaptability of technology to different agricultural *** purpose of this paper is to provide a comprehensive analysis of visual recognition technology for researchers and practitioners in the field of agricul-tural rob
Neuromorphic computing extends beyond sequential processing modalities and outperforms traditional von Neumann architectures in implementing more complicated tasks, e.g., pattern processing, image recognition, and dec...
详细信息
Neuromorphic computing extends beyond sequential processing modalities and outperforms traditional von Neumann architectures in implementing more complicated tasks, e.g., pattern processing, image recognition, and decision making. It features parallel interconnected neural networks, high fault tolerance, robustness, autonomous learning capability, and ultralow energy dissipation. The algorithms of artificial neural network (ANN) have also been widely used because of their facile self-organization and self-learning capabilities, which mimic those of the human brain. To some extent, ANN reflects several basic functions of the human brain and can be efficiently integrated into neuromorphic devices to perform neuromorphic computations. This review highlights recent advances in neuromorphic devices assisted by machinelearningalgorithms. First, the basic structure of simple neuron models inspired by biological neurons and the information processing in simple neural networks are particularly discussed. Second, the fabrication and research progress of neuromorphic devices are presented regarding to materials and structures. Furthermore, the fabrication of neuromorphic devices, including stand-alone neuromorphic devices, neuromorphic device arrays, and integrated neuromorphic systems, is discussed and demonstrated with reference to some respective studies. The applications of neuromorphic devices assisted by machinelearningalgorithms in different fields are categorized and investigated. Finally, perspectives, suggestions, and potential solutions to the current challenges of neuromorphic devices are provided. The review discusses the basic structure of simple neuron models inspired by biological neurons and how they process information in simple neural networks, laying the foundation for neuromorphic device *** progress in the fabrication of neuromorphic devices is highlighted, focusing on advancements in materials, structures, and the development of st
Single-pixel cameras are an effective solution for imaging beyond the visible spectrum, where traditional CMOS/CCD cameras face challenges. When combined with machinelearning, they can analyze images quickly enough f...
详细信息
Single-pixel cameras are an effective solution for imaging beyond the visible spectrum, where traditional CMOS/CCD cameras face challenges. When combined with machinelearning, they can analyze images quickly enough for practical applications. Solving the problem of high-dimensional single-pixel visualization can potentially be accelerated via quantum machinelearning, thereby expanding the range of practical problems. In this work, we simulated a single-pixel imaging experiment using Hadamard basis patterns, where images from the MNIST handwritten digit dataset and FashionMNIST items of clothing dataset were used as objects. There were selected 64 measurements with maximum variance (6% of the number of pixels in the image). We created algorithms for classifying and reconstructing images based on these measurements using classical fully-connected neural networks and parameterized quantum circuits. Classical and quantum classifiers showed the best accuracies of 96% and 95% for MNIST and 84% and 81% for FashionMNIST, respectively, after 6 training epochs, which is a quite competitive result. In the area of intersection by the number of parameters of the quantum and classical classifiers, the quantum demonstrates results no worse than the classical one, even better by a value of about 1-3%. image reconstruction was also demonstrated using classical and quantum neural networks after 10 training epochs;the best structural similarity index measure values were 0.76 and 0.26 for MNIST and 0.73 and 0.22 for FashionMNIST, respectively, which indicates that the problem in such a formulation turned out to be too difficult for quantum neural networks in such a configuration for now.
As advancements in agricultural technology unfold, machinelearning and deep learning approaches are gaining interest in robust plant disease identification. Early disease detection, integral to agricultural productiv...
详细信息
As advancements in agricultural technology unfold, machinelearning and deep learning approaches are gaining interest in robust plant disease identification. Early disease detection, integral to agricultural productivity, has propelled innovations across all phases of detection. This survey paper provides a meticulous examination of plant disease detection systems, elucidating data collection methodologies and underscoring the pivotal role of datasets in model training. The narrative navigates through the complex areas of data and imageprocessing techniques, segueing into an exploration of various segmentation methods. The survey emphasizes the importance of feature extraction and selection techniques, illustrating their efficacy in increasing classification accuracy. It examines the classification process, embracing both traditional machinelearning and avant-garde deep learning methods, with a particular spotlight on Convolutional neural networks (CNNs). The study examines over one hundred seminal papers, anatomizing their dataset utilizations, feature considerations, and classification strategies. Overall, the paper contemplates the challenges permeating this vibrant field, addressing critical issues such as dataset diversity, model generalization, and real-world applicability. Note to Practitioners-To ensure crop health and yield, timely and precise plant disease detection is crucial. Our research, titled "Advances And Challenges in Plant Disease Detection: A Comprehensive Survey of machine and Deep learning Approaches," examines the critical role of datasets, advanced imageprocessing, and segmentation techniques in disease detection. This paper presents practitioners with a guide to the latest techniques for enhanced disease detection by emphasizing the significance of feature extraction and highlighting the capabilities of convolutional neural networks (CNNs). By understanding the highlighted challenges, such as dataset diversity and model generalization, in
Convolutional neural networks (CNNs) are among the most promising algorithms, outperforming traditional methods in classification tasks with superior accuracy. They have been widely applied across various deep learnin...
详细信息
Convolutional neural networks (CNNs) are among the most promising algorithms, outperforming traditional methods in classification tasks with superior accuracy. They have been widely applied across various deep learning domains, including computer vision, speech recognition, imageprocessing, and object detection. However, many CNNs require substantial computational resources, particularly within their convolutional layers. As high-performance CNNs continue to evolve, their processing and memory requirements are also increasing. To address these challenges, this paper proposes an effective design methodology for accelerating CNN algorithms on Field-Programmable Gate Array (FPGA) hardware architectures. The proposed methodology introduces a novel approach for accelerating CNN algorithms using FPGAs, addressing the significant processing and memory demands associated with CNNs. The implementation is based on Open Computing Language (OpenCL), which provides rapid implementation flows. This approach was chosen for its efficiency in reducing development time and eliminating the need to manually write hardware description language (HDL) code. The MNIST and the CIFAR-10 datasets on the Xilinx ZYNQ 7000 device were used to evaluate our approach. Our method achieved a 97% recognition rate on MNIST and an 86% recognition rate on CIFAR-10. We compared the execution time of our accelerated CNN kernel on the FPGA with that of a single-core Central processing Unit (CPU). The experimental results demonstrate that our proposed design is 10 times faster than a standard CPU, validating its effectiveness. Our model optimizes power consumption and performance, exceeding previous studies in accuracy and efficiency. It is well suited for real-world applications that demand both precision and energy efficiency.
Deep learning has made important contributions to the development of medical image segmentation. Convolutional neural networks, as a crucial branch, have attracted strong attention from researchers. Through the tirele...
详细信息
Deep learning has made important contributions to the development of medical image segmentation. Convolutional neural networks, as a crucial branch, have attracted strong attention from researchers. Through the tireless efforts of numerous researchers, convolutional neural networks have yielded numerous outstanding algorithms for processing medical images. The ideas and architectures of these algorithms have also provided important inspiration for the development of later *** extensive experimentation, we have found that currently mainstream deep learningalgorithms are not always able to achieve ideal results when processing complex datasets and different types of datasets. These networks still have room for improvement in lesion localization and feature extraction. Therefore, we have created the dense multiscale attention and depth-supervised network (DmADs-Net).We use ResNet for feature extraction at different depths and create a Multi-scale Convolutional Feature Attention Block to improve the network's attention to weak feature information. The Local Feature Attention Block is created to enable enhanced local feature attention for high-level semantic information. In addition, in the feature fusion phase, a Feature Refinement and Fusion Block is created to enhance the fusion of different semantic *** validated the performance of the network using five datasets of varying sizes and types. Results from comparative experiments show that DmADs-Net outperformed mainstream networks. Ablation experiments further demonstrated the effectiveness of the created modules and the rationality of the network architecture.
To enhance public safety, crowd detection and prevention systems have essentially become a natural means to manage diverse crowded areas, such as urban settings, transportation hubs, and event venues. Recent systems t...
详细信息
To enhance public safety, crowd detection and prevention systems have essentially become a natural means to manage diverse crowded areas, such as urban settings, transportation hubs, and event venues. Recent systems take advantage of the synergy between machinelearning, data mining, and imageprocessing to extract/analyze features from crowded zones and recognize patterns and anomalies from the crowd behavior. Additionally, imageprocessing tools play a key role in real-time monitoring by analyzing video feeds to detect crowd density, flow direction, and identify potential risks like overcrowding or emergencies. However, most existing solutions focus on the detection phase and often overlook integrated error handling and robust decision-making frameworks to ensure accurate and actionable crowd prevention. Aiming to solve these issues, we take advantage of the prediction capabilities of machinelearning models and the analysis and clustering strengths of Formal Concept Analysis (FCA) chosen for its strong mathematical foundation and superior clustering capabilities compared to traditional methods, as highlighted in recent works such as K-means or hierarchical clustering. We used the first technique to extract useful knowledge from areas' produced images while mitigating potential error accumulation through modular error-checking mechanisms. A neural network is used to mark human bodies, determine the position of walking individuals, and predict crowd levels. Such information is, thereafter, inputted to the FCA-based decision system to ensure an explicit representation and modelling of crowd data, thanks to lattice structures. These latter's hierarchical view helped us identify the crowded areas and manage them as clustered zones, based on their common crowd information. We also define bottom-up parsing algorithms to recommend the suitable crowd prevention plan w.r.t. the crowd level. Experiments have successfully proved the ability of FCA to exclude low-crowd zones,
暂无评论