检索结果-内蒙古大学图书馆

2024 International Conference on image processing

作者： Xiao, Min Wang, Zi Guo, Jiefeng Guo, Di Qu, Xiaobo Xiamen Univ Inst Artificial Intelligence Xiamen Peoples R China Xiamen Univ Dept Elect Sci Fujian Prov Key Lab Plasma & Magnet Resonance Xiamen Peoples R China Xiamen Univ Dept Microelect & Integrated Circuit Xiamen Peoples R China Xiamen Univ Technol Sch Comp & Informat Engn Fujian Engn Res Ctr Med Data Min & Applicat Xiamen Peoples R China

ISBN: (纸本)9798350349405;9798350349399

Magnetic resonance imaging (MRI) plays a pivotal role in modern medical diagnosis yet is often hindered by the long imaging time. MRI imaging can be accelerated through undersampling, but the introduced aliasing artifacts should be removed during image reconstruction. While deep learning reconstruction methods excel at image de-aliasing, they may yield suboptimal results when training sampling settings differ from those at the time of reconstruction. To decouple from specific sampling settings, we propose using synthetic data to generate a substantial training dataset and pre-train a 1D deep denoiser. We then integrate the trained deep denoiser into the iterative reconstruction process as a replacement for the approximation operator within the deep plug-and-play framework. In vivo results indicate that the proposed method exhibits robust and visually appealing image reconstruction when there is a mismatch between the training and reconstruction undersampling settings, such as different undersampling patterns and sampling rates.

关键词： Magnetic resonance imaging image reconstruction deep learning synthetic data

来源：评论

学校读者我要写书评

暂无评论

Femtojoule optical nonlinearity for deep learning with incoherent illumination

引用

SCIENCE ADVANCES 2025年第5期11卷 eads4224页

作者： Feng, Qixin Uzundal, Can B. Guo, Ruihan Sanborn, Collin Qi, Ruishi Xie, Jingxu Zhang, Jianing Wu, Junqiao Wang, Feng Univ Calif Berkeley Dept Phys Berkeley CA 94720 USA Lawrence Berkeley Natl Lab Mat Sci Div Berkeley CA 94720 USA Univ Calif Berkeley Dept Chem Berkeley CA 94720 USA Univ Calif Berkeley Dept Mat Sci & Engn Berkeley CA 94720 USA Univ Calif Berkeley Grad Grp Appl Sci & Technol Berkeley CA 94720 USA ShanghaiTech Univ Sch Phys Sci & Technol Shanghai 201210 Peoples R China Univ Calif Berkeley Kavli Energy Nanosci Inst Berkeley CA 94720 USA

Optical neural networks (ONNs) are a promising computational alternative for deep learning due to their inherent massive parallelism for linear operations. However, the development of energy-efficient and highly parallel optical nonlinearities, a critical component in ONNs, remains an outstanding challenge. Here, we introduce a nonlinear optical microdevice array (NOMA) compatible with incoherent illumination by integrating the liquid crystal cell with silicon photodiodes at the single-pixel level. We fabricate NOMA with more than half a million pixels, each functioning as an optical analog of the rectified linear unit at ultralow switching energy down to 100 femtojoules per pixel. With NOMA, we demonstrate an optical multilayer neural network. Our work holds promise for large-scale and low-power deep ONNs, computer vision, and real-time optical image processing.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Integrating different detection techniques and data analysis methods for comprehensive food authenticity verification

引用

FOOD CHEMISTRY 2025年第Pt 4期463卷 141471页

作者： Liang, Chuxue Xu, Zhaoxin Liu, Pei Guo, Sheng Xiao, Ping Duan, Jin-ao Nanjing Univ Chinese Med Jiangsu Collaborat Innovat Ctr Chinese Med Resourc Natl & Local Collaborat Engn Ctr Chinese Med Resou Nanjing 210023 Peoples R China Nanjing Univ Chinese Med Jiangsu Key Lab High Technol Res TCM Formulae Nanjing 210023 Peoples R China

Traditional food testing methods, primarily confined to laboratory settings, are increasingly inadequate to detect covert food adulteration techniques. Hence, a crucial review of recent technological strides to combat food fraud is essential. This comprehensive analysis explores state-of-the-art technologies in food analysis, accentuating the pivotal role of sophisticated data processing methods and the amalgamation of diverse technologies in enhancing food authenticity testing. The paper assesses the merits and drawbacks of distinct data processing techniques and explores their potential synergies. The future of food authentication hinges on the integration of portable smart detection devices with mobile applications for real-time food analysis, including miniaturized spectrometers and portable sensors. This integration, coupled with advanced machine learning and deep learning for robust model construction, promises to achieve real-time, on-site food detection. Moreover, effective data processing, encompassing preprocessing, chemometrics, and regression analysis, remains indispensable for precise food authentication.

关键词： Food authenticity Spectroscopy Sensors DNA-based technology Mass spectrometry Machine learning

来源：评论

学校读者我要写书评

暂无评论

Robust Point Tracking Based On image Matching And Machine learning On Video images Taken from Fast Moving Camera 32

Robust Point Tracking Based On Image Matching And Machine Le...

引用

32nd IEEE Signal processing and Communications Applications Conference (SIU)

作者： Guven, Ali Yetik, Imam Samil TOBB Ekon & Teknol Univ Elekt & Elekt Muhendisligi Bolumu Ankara Turkiye

ISBN: (纸本)9798350388978;9798350388961

Recently, providing real-time navigation of unmanned aerial vehicles independent of global positioning systems has become of great importance. The state-of-the-art methods based on deep learning, which give good results in certain datasets, and the existing methods can not provide real-time and good solutions on images with dynamic and fast moving. Moreover, the methods, were developed so far, were focused on object-based tracking algorithms. In this paper, the tracking of the points belonging to the target pattern, found by image matching, was performed with the machine learning model we developed for 10 sequential video images. The features extracted for the machine learning model are: (i) the change between the points of the previous image and the image before that, (ii) the points of interest in the previous image, (iii) the changes found with the homography matrix between sequential images. It was experimentally shown that, point tracking can be achieved with the least error, on avarage about 23 pixels for a 2 mega-pixel resolution image, among the algorithms in the literature that can process more than 30 images per second in a CPU environment of 2 GHz or above.

关键词： image matching image homography image-object tracking image-point tracking apattern detection machine learning image processing feature extraction

来源：评论

学校读者我要写书评

暂无评论

Optimizing IoT Video Data: Dimensionality Reduction for Efficient deep learning on Edge Computing

引用

FUTURE INTERNET 2025年第2期17卷 53-53页

作者： Ortiz-Perez, David Ruiz-Ponce, Pablo Mulero-Perez, David Benavent-Lledo, Manuel Rodriguez-Juan, Javier Hernandez-Lopez, Hugo Iarovikov, Anatoli Krco, Srdjan Nedic, Daliborka Vukobratovic, Dejan Garcia-Rodriguez, Jose Univ Alicante Dept Comp Sci & Technol Alicante 03690 Spain DunavNet DOO Bulevar Oslobodenja 133-2 Novi Sad 21000 Serbia Univ Novi Sad Fac Tech Sci Trg Dositeja Obradovica 6 Novi Sad 21000 Serbia

The rapid loss of biodiversity significantly impacts birds' environments and behaviors, highlighting the importance of analyzing bird behavior for ecological insights. With the growing adoption of Machine learning (ML) algorithms in the Internet of Things (IoT) domain, edge computing has become essential to ensure data privacy and enable real-time predictions by processing high-dimensional data, such as video streams, efficiently. This paper introduces a set of dimensionality reduction techniques tailored for video sequences based on cutting-edge methods for this data representation. These methods drastically compress video data, reducing bandwidth and storage requirements while enabling the creation of compact ML models with faster inference speeds. Comprehensive experiments on bird behavior classification in rural environments demonstrate the effectiveness of the proposed techniques. The experiments incorporate state-of-the-art deep learning techniques, including pre-trained video vision models, Autoencoders, and single-frame feature extraction. These methods demonstrated superior performance to the baseline, achieving up to a 6000-fold reduction in data size while reaching a classification accuracy of 60.7% on the Visual WetlandBirds Dataset and obtaining state-of-the-art performance on this dataset. These findings underline the potential of using dimensionality reduction to enhance the scalability and efficiency of bird behavior analysis.

关键词： deep learning spatio-temporal dimensionality reduction bird behavior classification edge computing video processing

来源：评论

学校读者我要写书评

暂无评论

Semi-supervised learning for real-time anomaly detection in pulsed transfer wire arc additive manufacturing

引用

JOURNAL OF MANUFACTURING PROCESSES 2024年 128卷 84-97页

作者： Mattera, Giulio Polden, Joseph Caggiano, Alessandra Nele, Luigi Pan, Zengxi Norrish, John Univ Naples Federico II Dept Chem Mat & Ind Prod Engn Naples Italy Univ Wollongong Sch Mech Mat Mechatron & Biomed Engn Wollongong NSW Australia Univ Naples Federico II Fraunhofer Joint Lab Excellence Adv Prod Technol F Naples Italy Univ Naples Federico II Ctr Adv Metrol Technol Serv CeSMA Naples Italy

In Wire-Arc Additive Manufacturing (WAAM), ensuring the quality and integrity of components is of key importance and performing anomaly detection during production is an efficient strategy for preventing defects and maintaining an acceptable quality with lower costs compared to the development of complex supervised learning techniques. This paper presents an approach based on semi-supervised learning for real-time anomaly detection in the production of Inconel 718 components via the Pulsed Transfer WAAM process. The workflow is based on the use of training data obtained by employing established process parameters, similar to what is done in Welding Procedure Qualification Records (WPQRs), to develop a semi-supervised anomaly detection application. The proposed approach involves depositing material using the already established process parameters and collecting the corresponding welding data in terms of welding current and voltage sensor signals. The collected data are then employed to train an unsupervised algorithm by using solely good deposition data to learn hidden complex patterns of normality and hence detect anomalies based on deviations from these patterns. With this aim, global and local features are extracted from the welding current and voltage sensor signals in both time and time-frequency domains via Wavelet Transform technique and a deep learning approach based on a residual convolutional autoencoder. To showcase the effectiveness of the proposed approach, a case study involving wire arc additive manufacturing of Inconel 718 components was conducted. The proposed algorithm was tested and achieved an F-score of 0.895, representing a 30 % improvement over state-of-the-art methods that rely on time domain features for anomaly detection. This research work contributes to the improvement of anomaly detection methodologies, offering substantial advantages over alternative techniques for quality control and reliability of additive manufacturing processes such

关键词： Wire arc additive manufacturing Anomaly detection Machine learning Frequency domain image processing

来源：评论

学校读者我要写书评

暂无评论

Research on Hardware Acceleration of Traffic Sign Recognition Based on Spiking Neural Network and FPGA Platform

引用

IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS 2025年第2期33卷 499-511页

作者： Chen, Huarun Liu, Yijun Ye, Wujian Ye, Jialiang Chen, Yuehai Chen, Shaozhen Han, Chao Guangdong Univ Technol Sch Integrated Circuits Guangzhou 510006 Peoples R China Guangdong Univ Technol Res Inst IC Innovat RIICI Guangzhou 510006 Peoples R China

Most of the existing methods for traffic sign recognition exploited deep learning technology such as convolutional neural networks (CNNs) to achieve a breakthrough in detection accuracy;however, due to the large number of CNN's parameters, there are problems in practical applications such as high power consumption, large calculation, and slow speed. Compared with CNN, a spiking neural network (SNN) can effectively simulate the information processing mechanism of biological brain, with stronger parallel processing capability, better sparsity, and real-time performance. Thus, we design and realize a novel traffic sign recognition system called SNN on FPGA-traffic sign recognition system (SFPGA-TSRS) based on spiking CNN (SCNN) and FPGA platform. Specifically, to improve the recognition accuracy, a traffic sign recognition model spatial attention SCNN (SA-SCNN) is proposed by combining LIF/IF neurons based SCNN with SA mechanism;and to accelerate the model inference, a neuron module is implemented with high performance, and an input coding module is designed as the input layer of the recognition model. The experiments show that compared with existing systems, the proposed SFPGA-TSRS can efficiently support the deployment of SCNN models, with a higher recognition accuracy of 99.22%, a faster frame rate of 66.38 frames per second (FPS), and lower power consumption of 1.423 W on the GTSRB dataset.

关键词： Accuracy Neurons Field programmable gate arrays Power demand image recognition real-time systems Parallel processing Biological system modeling image coding Encoding Attention fusion field-programmable gate array (FPGA) platform input coding optimization spiking convolutional neural network (SCNN) traffic sign recognition

来源：评论

学校读者我要写书评

暂无评论

A novel framework using 3D-CNN and BiLSTM model with dynamic learning rate scheduler for visual speech recognition

引用

SIGNAL image AND VIDEO processing 2024年第6-7期18卷 5433-5448页

作者： Chandrabanshi, Vishnu Domnic, S. Natl Inst Technol Dept Comp Applicat Tiruchirappalli 620015 Tamil Nadu India

Visual Speech Recognition (VSR) is an appealing technology for predicting and analyzing spoken language based on lip movements. Previous research in this area has primarily concentrated on leveraging both audio and visual cues to achieve enhanced accuracy in speech recognition. However, existing solutions encounter significant limitations, including inadequate training data, variations in speech patterns, and similar homophones, which need more comprehensive feature representation to improve accuracy. This article presents a novel deep learning model for performing word level VSR. In this study, we have introduced a dynamic learning rate scheduler to adapt the learning parameter during model training. Additionally, we employ an optimized Three-Dimensional Convolution Neural Network for extracting spatio-temporal features. To enhance context processing and ensure accurate mapping of input sequences to output sequences, we combine Bidirectional Long Short Term Memory with the CTC loss function. We have utilized the GRID dataset to assess word-level metrics, including Word Error Rate (WER) and Word Recognition Rate (WRR). The model achieves 1.11% WER and 98.89% WRR, respectively, for overlapped speakers. This result demonstrates that our strategy outperforms and is more effective than existing VSR methods. Practical Implications - The proposed work aims to elevate the accuracy of VSR, facilitating its seamless integration into real-time applications. The VSR model finds applications in liveness detection for person authentication, improving password security by not relying on written or spoken passcodes, underwater communications and aiding individuals with hearing and speech impairments in the medical field.

关键词： 3D-CNN BiLSTM Visual speech recognition LRS deep learning

来源：评论

学校读者我要写书评

暂无评论

Efficient convolutional neural networks on Raspberry Pi for image classification

引用

Journal of real-time image processing 2023年第2期20卷 21页

作者： Ju, Rui-Yang Lin, Ting-Yu Jian, Jia-Hao Chiang, Jen-Shiun Department of Electrical and Computer Engineering Tamkang University No. 151 Yingzhuan Rd. Tamsui Dist. New Taipei City251301 Taiwan Department of Engineering Science National Cheng Kung University No. 1 University Rd. East Dist. Tainan City70101 Taiwan

With the good performance of deep learning in the field of computer vision (CV), the convolutional neural network (CNN) architectures have become main backbones of image recognition tasks. With the widespread use of mobile devices, neural network models based on platforms with low computing power are gradually being paid attention. However, due to the limitation of computing power, deep learning algorithms are usually not available on mobile devices. This paper proposes a lightweight convolutional neural network TripleNet, which can operate easily on Raspberry Pi. Adopted from the concept of block connections in ThreshNet, the newly proposed network model compresses and accelerates the network model, reduces the amount of parameters of the network, and shortens the inference time of each image while ensuring the accuracy. Our proposed TripleNet and other State-of-the-Art (SOTA) neural networks perform image classification experiments with the CIFAR-10 and SVHN datasets on Raspberry Pi. The experimental results show that, compared with GhostNet, MobileNet, ThreshNet, EfficientNet, and HarDNet, the inference time of TripleNet per image is shortened by 15%, 16%, 17%, 24%, and 30%, respectively. The detail codes of this work are available at https://***/RuiyangJu/TripleNet. © 2023, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： image classification

来源：评论

学校读者我要写书评

暂无评论

High-Efficiency FPGA Implementation of deep learning Models for Colorectal Tumor image Analysis 2

High-Efficiency FPGA Implementation of Deep Learning Models ...

引用

2nd IEEE Global Conference on Communications and Information Technologies, GCCIT 2024

作者： Asha, S. Chandru, R. Rohit, S. Saveetha Engineering College Department of Ece Tamil Nadu Chennai India

ISBN: (纸本)9798350388916

To meet the demand for real-time, energy-efficient solutions in medical diagnostics, this research offers a high-efficiency FPGA implementation of a convolutional neural network (CNN) for colorectal tumor picture processing. Loop unrolling, pipelining, and effective memory use were among the methods used in the creation of the CNN model, which had an optimized system architecture. With a lowered latency of 10.2 milliseconds and power usage of 4.8 watts, the model achieved an accuracy of 98.5% during training and testing on the TCGA Colorectal Tumor Dataset. With a throughput of 120 frames per second, the implementation from 2021 to 2023 significantly outperformed earlier ones. These findings highlight the possibility of FPGA-based deep learning models to provide real-time, high-performance analysis at a lower energy cost. According to the results, this strategy not only improves diagnostic precision but also provides a practical, scalable solution for clinical setting deployment, opening the door for a wider use of FPGA-based systems in medical image processing. © 2024 IEEE.

关键词： Convolutional neural networks

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：