检索结果-内蒙古大学图书馆

Design Space Exploration for Edge Machine learning Featured by MathWorks FPGA DL Processor: A Survey

IEEE ACCESS 2024年 12卷 9418-9439页

作者： Bertazzoni, Stefano Canese, Lorenzo Cardarilli, Gian Carlo Di Nunzio, Luca Fazzolari, Rocco Re, Marco Spano, Sergio Tor Vergata Univ Rome Dept Elect Engn I-00133 Rome Italy

This paper proposes a Design Space Exploration for Edge machine learning through the utilization of the novel MathWorks FPGA deep learning Processor IP, featured in the HDL deep learning toolbox. With the ever-increasing demand for real-time machine learning applications, there is a critical need for efficient and low-latency hardware solutions that can operate at the edge of the network, in close proximity to the data source. The HDL deep learning toolbox provides a flexible and customizable platform for deploying deep learning models on FPGAs, enabling effective inference acceleration for embedded IoT applications. In this study, our primary focus lies in investigating the impact of parallel processing elements on the performance and resource utilization of the FPGA-based processor. By analyzing the trade-offs between accuracy, speed, energy efficiency, and hardware resource utilization, we aim to gain valuable insights into making optimal design choices for FPGA-based implementations. Our evaluation is conducted on the AMD-Xilinx ZC706 development board, which serves as the target device for our experiments. We consider all the compatible Convolutional Neural Networks available within the HDL deep learning toolbox to comprehensively assess the performances.

关键词： Convolutional neural networks deep learning design space exploration edge machine learning embedded FPGA IoT machine learning

来源：评论

学校读者我要写书评

暂无评论

time-attentive fusion network: An efficient model for online detection of action start

引用

IET image processing 2024年第7期18卷 1892-1902页

作者： Hu, Xuejiao Wang, Shijie Li, Ming Li, Yang Du, Sidan Nanjing Univ Sch Elect Sci & Engn Nanjing Peoples R China

Online detection of action start is a significant and challenging task that requires prompt identification of action start positions and corresponding categories within streaming videos. This task presents challenges due to data imbalance, similarity in boundary content, and real-time detection requirements. Here, a novel time-Attentive Fusion Network is introduced to address the requirements of improved action detection accuracy and operational efficiency. The time-attentive fusion module is proposed, which consists of long-term memory attention and the fusion feature learning mechanism, to improve spatial-temporal feature learning. The temporal memory attention mechanism captures more effective temporal dependencies by employing weighted linear attention. The fusion feature learning mechanism facilitates the incorporation of current moment action information with historical data, thus enhancing the representation. The proposed method exhibits linear complexity and parallelism, enabling rapid training and inference speed. This method is evaluated on two challenging datasets: THUMOS'14 and ActivityNet v1.3. The experimental results demonstrate that the proposed method significantly outperforms existing state-of-the-art methods in terms of both detection accuracy and inference speed. Here, a novel time-Attentive Fusion Network (TAF-Net) is introduced to address the requirements of improved action detection accuracy and operational efficiency in the task of online detection of action start. The proposed model not only learns valuable sequence information for precise detection but its linear computational complexity and parallelism also contribute to a faster inference speed. image

关键词： computer vision feature extraction image processing video signal processing

来源：评论

学校读者我要写书评

暂无评论

A³R-Net: adaptive attention aggregation residual network for sparse DOA estimation

引用

SIGNAL image AND VIDEO processing 2024年第3期18卷 2939-2949页

作者： Xu, Qihui Huang, Qinghua Shanghai Univ Sch Commun & Informat Engn Shanghai 200444 Peoples R China

In this paper, a unified deep learning framework is developed for high-precision direction-of-arrival (DOA) estimation. Unlike previous methods that divide the real and imaginary parts of complex-valued sparse problem into two separate input channels, a real-valued transformation is adopted to encode the correlation between them. Then, a novel adaptive attention aggregation residual network (A(3)R-Net) is designed to overcome the challenges in the case of low signal-to-noise ratios or small inter-signal angle separations. First, to alleviate the gradient disappearance and gradient explosion caused by network deepening, a residual learning strategy is introduced to construct a deep estimation network that learns the inverse mapping from the array measurement vector to the original spatial spectrum. Second, since the feature fusion method via simple summation in the shortcut connection ignores the inconsistency on the scale and semantic of features, an adaptive attention aggregation module (A(3)M) with adaptive channel context aggregators is proposed to capture multi-scale channel contexts and generate element-wise fusion weights. Finally, a dilated convolution with a broader receptive field is embedded into the channel context aggregator to learn wider local cross-channel association. Extensive simulation results demonstrate the superiority and robustness of the proposed method compared with other state-of-the-art methods.

关键词： Direction-of-arrival (DOA) estimation deep learning Feature fusion Attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Research on an Innovative Algorithm for Optimizing Intelligent image Data Systems Based on deep learning 7

Research on an Innovative Algorithm for Optimizing Intellige...

引用

7th International Conference on Automation Electronics and Electrical Engineering

作者： Dong, Xiaonan Song, Yingbin Wuhan Technol & Business Univ Wuhan 430065 Hubei Peoples R China

ISBN: (纸本)9798350377040;9798350377033

This paper proposes an innovative algorithm for optimizing intelligent image data systems based on deep learning. The algorithm combines image feature extraction, data preprocessing and efficient optimization strategies to improve the performance and accuracy of image data processing systems. First, by designing a deep CNN architecture, important features in the image are extracted to achieve efficient completion of image recognition and classification tasks. Subsequently, a new multi-level data processing method is proposed, which can optimize image data at different levels, thereby improving processing speed and reducing noise interference. Through a series of simulation experiments, the results show that the image classification accuracy of the algorithm is improved by about 12%, from 85.6% of the traditional method to 97.3%. In addition, the processing efficiency is improved by about 20%, the data processing time is reduced from 2.5 seconds of the traditional method to 2 seconds, and the stability of the system is significantly enhanced by introducing optimization strategies, and the stability is improved by about 18%. The optimized algorithm shows significant advantages in both accuracy and efficiency, meeting the needs of efficient intelligent image processing systems.

关键词： deep learning intelligent image processing data processing

来源：评论

学校读者我要写书评

暂无评论

D-TLDetector: Advancing Traffic Light Detection With a Lightweight deep learning Model

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2025年第3期26卷 3917-3933页

作者： Huang, Yinjie Wang, Fuyuan Yancheng Inst Technol Sch Mech Engn Yancheng 224051 Peoples R China

Traffic signal light detection poses significant challenges in the intelligent driving sector, with high precision and efficiency being crucial for system safety. Advances in deep learning have led to significant improvements in image object detection. However, existing methods continue to struggle with balancing detection speed and accuracy. We propose a lightweight model for traffic light detection that uses a streamlined backbone network and a Low-GD neck architecture. The model's backbone employs structured reparameterization and lightweight Vision Transformers, using multi-branch and Feed-Forward Network structures to boost informational richness and positional awareness, respectively. The Neck network utilizes the Low-GD structure to enhance the aggregation and integration of multi-scale features, reducing information loss during cross-layer exchanges. We introduce a data augmentation strategy using Stable Diffusion to expand our traffic light dataset in complex weather conditions like fog, rain, and snow, improving model generalization. Our method excels on the YCTL2024 traffic light dataset, achieving a detection speed of 135 FPS and 98.23% accuracy, with only 1.3M model parameters. Testing on the Bosch Small Traffic Lights Dataset confirms the method's strong generalization capabilities. This suggests that our proposed method can effectively provide accurate and real-time traffic light detection.

关键词： Feature extraction Accuracy Computational modeling Generative adversarial networks Neck image color analysis Meteorology deep learning Data augmentation Training Intelligent transportation system traffic light detection deep learning lightweight network stable diffusion

来源：评论

学校读者我要写书评

暂无评论

Approach to Smart Mobility Intelligent Traffic Signal System based on Distributed deep Reinforcement learning

引用

IEIE Transactions on Smart processing and Computing 2024年第1期13卷 89-95页

作者： Lee, Young-Sik Department of Software Kyungdong University Yangju11458 Korea Republic of

Smart mobility intelligent traffic services have become critical in intelligent transportation systems (ITS). This involves using advanced sensors and controllers and the ability to respond to real-time traffic situations at intersections, alleviate congestion, and generate policies to prevent traffic jams. deep reinforcement learning (DRL) provides a natural framework for processing tasks. In DRL, each intersection can control itself and coordinate with neighbors to achieve optimal network-wide policies. On the other hand, comparing approaches remains a challenging task due to the existence of numerous possible configurations. This research performs a critical comparison of various traffic controllers in the literature. Hence, using a nonlinear approximator for coordination mechanisms and enhancing observability at each intersection are key performance drivers. Copyrights © 2024 The Institute of Electronics and Information Engineers.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

PhagoStat a scalable and interpretable end to end framework for efficient quantification of cell phagocytosis in neurodegenerative disease studies

引用

SCIENTIFIC REPORTS 2024年第1期14卷 1-30页

作者： Ounissi, Mehdi Latouche, Morwena Racoceanu, Daniel Sorbonne Univ Inria AP HP CNRSParis Brain Inst ICMInserm F-75013 Paris France Sorbonne Univ Inst Cerveau AP HP CNRSInsermICM F-75013 Paris France PSL Res Univ EPHE Paris France

Quantifying the phagocytosis of dynamic, unstained cells is essential for evaluating neurodegenerative diseases. However, measuring rapid cell interactions and distinguishing cells from background make this task very challenging when processing time-lapse phase-contrast video microscopy. In this study, we introduce an end-to-end, scalable, and versatile real-time framework for quantifying and analyzing phagocytic activity. Our proposed pipeline is able to process large data-sets and includes a data quality verification module to counteract potential perturbations such as microscope movements and frame blurring. We also propose an explainable cell segmentation module to improve the interpretability of deep learning methods compared to black-box algorithms. This includes two interpretable deep learning capabilities: visual explanation and model simplification. We demonstrate that interpretability in deep learning is not the opposite of high performance, by additionally providing essential deep learning algorithm optimization insights and solutions. Besides, incorporating interpretable modules results in an efficient architecture design and optimized execution time. We apply this pipeline to quantify and analyze microglial cell phagocytosis in frontotemporal dementia (FTD) and obtain statistically reliable results showing that FTD mutant cells are larger and more aggressive than control cells. The method has been tested and validated on several public benchmarks by generating state-of-the art performances. To stimulate translational approaches and future studies, we release an open-source end-to-end pipeline and a unique microglial cells phagocytosis dataset for immune system characterization in neurodegenerative diseases research. This pipeline and the associated dataset will consistently crystallize future advances in this field, promoting the development of efficient and effective interpretable algorithms dedicated to the critical domain of neurodegenerative disease

关键词： Explainable artificial intelligence (XAI) deep learning (DL) Interpretability Scalability Phase-contrast video microscopy Neuro-phagocytosis Neurodegenerative disease image processing

来源：评论

学校读者我要写书评

暂无评论

Fast-DSAGCN: Enhancing semantic segmentation with multifaceted attention mechanisms

引用

NEUROCOMPUTING 2024年 587卷

作者： Rashid, Khawaja Iftekhar Yang, Chenhui Huang, Chenxi Xiamen Univ Sch Informat Xiamen 361005 Peoples R China

real-time semantic segmentation provides precise insights into dynamic street environments for autonomous driving, traffic control, and urban planning. However, state-of-the-art models following attention mechanisms and deep convolutional neural networks have improved semantic segmentation at the cost of complex architectures and high computation complexity. The study aims to mitigate the presence of gridding artifacts and enhance semantic segmentation performance. In addition, we propose a multi-level downsampling approach before employing the depth-wise split separable global convolution with the bottleneck to achieve a trade-off between accuracy and inference time. The spatial attention module used in this study effectively keeps lowlevel spatial characteristics, enhancing the accuracy of localization, robustness against disturbances, processing efficiency, and the ability to handle occlusions. Thorough tests of the Cityscapes and CamVid datasets available for public access indicate that the model presented is capable of efficiently processing high-resolution photos in real time, resulting in exceptional performance. The model has achieved an accuracy of 72.3% on the cityscapes dataset and 72.7% on the CamVid dataset.

关键词： deep learning Linear bottleneck Neural network Semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

Advanced analysis of disintegrating pharmaceutical compacts using deep learning-based segmentation of time-resolved micro-tomography images

引用

HELIYON 2024年第4期10卷 e26025页

作者： Waldner, Samuel Wendelspiess, Erwin Detampel, Pascal Schlepuetz, Christian M. Huwyler, Jorg Puchkov, Maxim Univ Basel Dept Pharmaceut Sci Div Pharmaceut Technol Klingelbergstr 50 CH-4056 Basel Switzerland Paul Scherrer Inst Swiss Light Source CH-5232 Villigen Switzerland

The mechanism governing pharmaceutical tablet disintegration is far from fully understood. Despite the importance of controlling a formulation's disintegration process to maximize the active pharmaceutical ingredient's bioavailability and ensure predictable and consistent release profiles, the current understanding of the process is based on indirect or superficial measurements. Formulation science could, therefore, additionally deepen the understanding of the fundamental physical principles governing disintegration based on direct observations of the process. We aim to help bridge the gap by generating a series of time-resolved X-ray microcomputed tomography (mu CT) images capturing volumetric images of a broad range of minitablet formulations undergoing disintegration. Automated image segmentation was a prerequisite to overcoming the challenges of analyzing multiple time series of heterogeneous tomographic images at high magnification. We devised and trained a convolutional neural network (CNN) based on the U-Net architecture for autonomous, rapid, and consistent image segmentation. We created our own mu CT data reconstruction pipeline and parameterized it to deliver image quality optimal for our CNN-based segmentation. Our approach enabled us to visualize the internal microstructures of the tablets during disintegration and to extract parameters of disintegration kinetics from the time-resolved data. We determine by factor analysis the influence of the different formulation components on the disintegration process in terms of both qualitative and quantitative experimental responses. We relate our findings to known formulation component properties and established experimental results. Our direct imaging approach, enabled by deep learning-based image processing, delivers new insights into the disintegration mechanism of pharmaceutical tablets.

关键词： Disintegration Swelling Tablets time -resolved micro -computed tomography deep learning -based image segmentation

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Various CNN Architectures in Recognizing Objects in a Classification System 9

Comparative Analysis of Various CNN Architectures in Recogni...

引用

9th IEEE International Conference for Convergence in Technology, I2CT 2024

作者： Surve, Yash Pudari, Kshitija Bedade, Sonali Masanam, Balaji Durai Bhalerao, Kinnari Mhatre, Pratik Vidyalankar Institute of Technology Department of Electronics And Telecommunication Engineering Mumbai India

ISBN: (纸本)9798350394474

Object recognition, an essential technique in computer vision, enables machines to identify and understand real-time objects and environments based on input images. The main aim of this technology is to accurately recognize image features to enable accurate object recognition. With the rapid evolution of various machine learning (ML) and deep learning (DL) algorithms, the world of image processing and computer vision has witnessed significant growth. deep learning algorithms offer high accuracy by processing vast amounts of data, while machine learning algorithms provide flexibility in selecting the best combinations of features and classifiers for learning purposes. Convolutional Neural Network (CNN) is a widely used deep-learning model that is highly effective in classifying images. AlexNet, VGG16, ResNet and GoogleNet are some of the well-known CNN architectures used for object recognition. The paper proposes a comparative analysis of prediction accuracy between these pre-trained CNN models using transfer learning. The transfer learning approach, on the other hand, involves the fine-tuning of pre-trained models to enhance prediction accuracy in various image classification scenarios. In this paper, the CNN models are trained and tested on a dataset containing images of apples, oranges and bananas. Experimental results with a real dataset show that MobileNet V2 has the highest accuracy of 92.80%. © 2024 IEEE.

关键词： image classification

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：