检索结果-内蒙古大学图书馆

SPIE 12th International Symposium on Multispectral Image processing and Pattern Recognition, MIPPR 2023

作者： Ye, Mao Zhu, Wenjie Liu, Kewei Zhang, Xinyu Institude of Artificial Intelligence Huazhong University of Science and Technology Hubei Wuhan430074 China National Key Laboratory of Science & Technology on Multispectral Information Processing Huazhong University of Science & Technology Wuhan430074 China School of Artificial Intelligence & Automation Huazhong University of Science & Technology Wuhan430074 China

ISBN: (数字)9781510674929

ISBN: (纸本)9781510674912

Polarization is the asymmetry of a light wave's electric field vibration direction with respect to its propagation direction, as a crucial clue for describing the light, which is one of the most apparent signs that distinguishes transverse waves from longitudinal waves. The current mainstream polarimetric cameras typically adopt a division of focal plane imaging architecture, in which a metal wire-grid structure is coupled with a detector to acquire the polarization state of the target. However, the structure of the division of focal plane reduces effective spatial resolution of the detector and requires complex registration algorithms to achieve pixel-level polarization acquisition, which leads to high equipment costs and impedes the development and application of polarimetric cameras in industrial production. In this paper, a single-layer twisted nematic liquid crystal (TNLC) is inserted to the imaging optical path of traditional camera to realize the polarization acquisition, allowing for a cost-effective transition from traditional devices to polarization imaging systems. The imaging detector coupled with the TNLC device can acquire polarization states ranging from 10 to 90-degree by adjusting the external voltage applied on the LC device. Moreover, a polarization angle calibration method based on a single-layer TNLC is proposed in this paper, providing a reliable reference for electrically controlled polarization angle acquisition. The experimental results acquired by presented imaging system demonstrate the feasibility and effectiveness of the proposed method. The TNLC based imaging scheme presented in this paper provides a solid foundation for developing simple, low-cost, and efficient polarimetric cameras. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

Compressed Sensing Based on K-SVD for Brillouin optical Fiber Distributed Sensors

引用

IEEE SENSORS JOURNAL 2022年第16期22卷 16414-16421页

作者： Dong, Yong Yang, Ya-Nan Azad, Abul Kalam Yang, Zengsen Yu, Kuanglu Zhao, Shuang Beijing Key Lab Adv Informat Sci & Network Techno Beijing 100044 Peoples R China Beijing Jiaotong Univ Inst Informat Sci Sch Comp & Informat Technol Beijing 100044 Peoples R China Univ Dhaka Dept Elect & Elect Engn Dhaka 1000 Bangladesh China Household Elect Appliances Res Inst Beijing 100053 Peoples R China

The continuous operation of Brillouin optical fiber distributed sensors requires the acquisition and processing of large amount of data that imposes ample pressure on the data acquisition systems and storage devices. To overcome such limitations, the compressed sensing (CS) method based on K-Singular value decomposition (K-SVD) algorithm as transform is proposed in this paper for the accurate extraction of sensing information with less data acquired from Brillouin optical fiber distributed sensors without the hardware modification. In such algorithm, the parameters of the K-SVD dictionary are selected first through simulation and the signals are sampled and reconstructed successfully. Then, we apply the regularized orthogonal matching pursuit (ROMP) reconstruction algorithm to effectively reconstruct the experimental data obtained with different sampling rates, frequency steps and test temperatures. We also compare the performances of the CS method based on K-SVD with that based on discrete cosine transform (DCT), discrete Fourier transform (DFT) and principle component analysis (PCA) as transforms. The experimental results show that the proposed method can effectively reconstruct the experimental signals with much reduced amount of data as compared to other methods. The proposed method also provides satisfactory measurement accuracy in extracting sensing information. Therefore, the K-SVD based CS method can be an attractive alternative to make Brillouin optical fiber distributed sensors more suitable for faster operation.

关键词： Sensors optical sensors Sparse matrices Scattering Dictionaries Matching pursuit algorithms Transforms Brillouin backscattering compressed sensing distributed fiber sensor K-SVD ROMP

来源：评论

学校读者我要写书评

暂无评论

Optoelectronic artificial synapses based on copper (ii) phthalocyanine with modulated neuroplasticity

引用

JOURNAL OF MATERIALS SCIENCE-MATERIALS IN ELECTRONICS 2022年第23期33卷 18497-18506页

作者： Ji, Rongxue Li, Ruixiang Feng, Guangdi Lao, Jie Jiang, Chunli Tian, Bobo Luo, Chunhua Lin, Hechun Peng, Hui Duan, Chun-Gang East China Normal Univ Sch Phys & Elect Sci Dept Elect Key Lab Polar Mat & Devices MOE Shanghai 200241 Peoples R China Shanxi Univ Collaborat Innovat Ctr Extreme Opt Taiyuan 030006 Shanxi Peoples R China Chinese Acad Sci Shanghai Inst Tech Phys Natl Lab Infrared Phys Shanghai 200083 Peoples R China

Optoelectronic synapses have been attracting significant attention due to their important role in visual information processing. In this work, we fabricate an all-organic optoelectronic synaptic device with a double heterojunction structure of PEDOT:PSS/poly(vinylidene fluoridetrifluoroethylene) (P(VDF-TrFE))/copper (ii) phthalocyanine (CuPc) by a simple preparation process. The introduction of a dielectric P(VDF-TrFE) layer between PEDOT:PSS and CuPc layers benefits the trapping of charge carriers and slows down the electron-hole recombination rate. This two-terminal optoelectronic device is successfully applied to simulate synaptic functions of biological synapses by using optical pulses of 660 nm, including paired-pulse facilitation, spike-duration dependent plasticity, spike-rate dependent plasticity, spike-number dependent plasticity, and learning-experience behavior. Furthermore, the key characteristics of a nociceptor and the optical logic function of the "AND" and "OR" operations are also emulated. This work illustrates the potential of such device for constructing neuromorphic computing systems at the physical level.

关键词： Computation theory

来源：评论

学校读者我要写书评

暂无评论

Development of an Entry-Exit Decision Algorithm for Bus Passenger Counting Using Cameras 16

Development of an Entry-Exit Decision Algorithm for Bus Pass...

引用

2024 16th International Conference on Graphics and Image processing, ICGIP 2024

作者： Nishiura, Tsubasa Yokoyama, Soichiro Yamashita, Tomohisa Kawamura, Hidenori Sato, Yoshimi Hasegawa, Rei Hirasawa, Miyuki Faculty of Information Science and Technology Hokkaido University Kita 14 Nishi 9 Kita-ku Hokkaido Sapporo060-0814 Japan C’s Lab Co. Ltd. 3 Kita 1-jo Nishi 7-chome Chuo-ku Hokkaido Sapporo060-0001 Japan

ISBN: (数字)9781510688780

ISBN: (纸本)9781510688773

In recent years, entry-exit management technologies have gained attention due to the rise of smart cities and the proliferation of IoT devices. These technologies are used not only in indoor environments like offices and retail stores but also in outdoor settings such as public transportation. Particularly in public transport, there is growing demand for Origin-Destination (OD) data collection to better understand passenger movement. In this study, we developed a system to estimate the number of passengers boarding and alighting from buses by installing cameras at the bus doors and capturing video footage of passengers. The system utilizes person detection and tracking algorithms to estimate passenger trajectories and applies a deep neural network (DNN) model to classify whether a detected individual is boarding or alighting. We introduced a DNN model for binary classification to determine whether individuals are inside or outside the bus, using bounding box and time information. By performing a grid search, we identified optimal model parameters. The system was tested using both YOLOR and the latest YOLOv10 object detectors, and the MPNTrack person tracking algorithm. The combination of YOLOv10 and the DNN model achieved an error rate of 11.7% for the entrance camera, while YOLOR and the DNN model achieved an error rate of 0.803% for the exit camera, significantly improving the accuracy over previous methods. This study demonstrates that integrating object detection, tracking, and DNN-based classification improves the accuracy of passenger counting systems, which is crucial for the effective management of public transportation systems. © 2025 SPIE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

EM2-YOLO: Lightweight Remote Sensing Image Detection 5th

EM2-YOLO: Lightweight Remote Sensing Image Detection

引用

5th International Conference on 3D Imaging Technologies—Multidimensional Signal processing and Deep Learning, 3DIT-MSP and DL 2023

作者： Liu, Xinyi Song, Yaolian Wang, Can School of Information Engineering and Automation Kunming University of Science and Technology Yunnan650500 China

ISBN: (纸本)9789819751808

Remote sensing photographs have a wealth of information, and object detection methods are crucial in this. On mobile and embedded platforms, deep learning-based target identification algorithms are challenging to implement. Additionally, the issues of delayed target detection and poor accuracy of small target detection for remote sensing photos must be resolved immediately. To address these issues, we provide EM2-YOLOv4, a simple method for recognizing remote sensing images. MobileNetv2 acts as the backbone network to reduce model complexity, while depth-separable convolution enhances detection speed. Additionally, we introduce a feature enhancement module and an Efficient Channel Attention (ECA) module to improve model accuracy. To address the issue of unequal assignment of positive and negative samples, we incorporate Focal Loss into the loss function. Experimental demonstrate that the model can perform detection at 45 frames per second with an average accuracy of 94.11%. The model has been significantly reduced, as seen by the fact that it is only 48 MB in size, but it still maintains good detection accuracy and accelerates detection. The model is ideal for usage on embedded devices because of this feature. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： optical remote sensing

来源：评论

学校读者我要写书评

暂无评论

Implementation of dual-phase grating system phase contrast information extraction based on generative adversarial networks 4

Implementation of dual-phase grating system phase contrast i...

引用

4th International Conference on Laser, Optics, and Optoelectronic Technology, LOPET 2024

作者： Zeng, Jiacheng Huang, Jianheng Zhu, Minghui Li, Jiaqi Lei, Yaohu Liu, Xin Key Laboratory of Optoelectronic Devices and Systems of Ministry of Education and Guangdong Province College of Physics and Optoelectronic Engineering Shenzhen University Shenzhen China

ISBN: (纸本)9781510681903

In the dual-phase grating system of X-ray phase-contrast imaging, the reduction in structural requirements for the absorption grating relaxes the constraints on its aspect ratio and period, thereby broadening the applicability of X-ray phase-contrast imaging techniques in a wider market. Traditionally, the extraction of phase-contrast information primarily relies on phase-stepping methods or Fourier transform algorithms, which often introduce artifacts and blurring into the images. To achieve higher quality image restoration, this study introduces generative adversarial networks (GANs) for high-quality image reconstruction. Our approach uses ideal images as labels and images containing object stripe information as inputs, utilizing GANs for feature learning to facilitate the transformation from object stripe images to high-quality phase-contrast images. The network also employs transfer learning to process previously unseen object stripe images and generate corresponding phase-contrast images. This technique not only significantly enhances image resolution but also substantially reduces artifacts and blurring in the image processing, paving the way for high-precision demands in medical diagnostics and industrial inspection. © 2024 SPIE.

关键词： Image resolution

来源：评论

学校读者我要写书评

暂无评论

Webcam-Based optical Mark Reader for Grading Assessments

Webcam-Based Optical Mark Reader for Grading Assessments

引用

作者： Muhammad Fahad Zafar University of Windsor

学位级别：硕士

optical Mark Recognition (OMR) is a process of retrieving information from the forms where users put marks in particular locations. OMR technique has been improved significantly since its inception, moving from specialized hardware-based solutions to more adaptable and sophisticated software-based solutions using a scanner. Recently, webcam-based software solutions have emerged because of their affordability and ease of use. In this thesis, we propose an improved webcam-based OMR solution for grading the student tests given on a specifically formatted answer sheet. We apply various image processing techniques to achieve effective grading, resolving technical problems stemming from multiple factors like image orientation, skewness, and uneven illumination. In particular, we propose a new variant of the Otsu binarization technique for image segmentation to effectively separate the background and foreground. The proposed grading solution aims at maximizing the criteria for acceptable operations so that 100% accurate grading among those accepted ones can be achieved. The evaluation of the proposed method is two-fold: (i) the proposed variant of Otsu is compared with other binarization algorithms for its effectiveness in background separation; (ii) the grading system is compared with the state-of-the-art webcam-based systems where the individual contribution of each applied image processing technique to the grading enhancement will be identified.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Using Image processing and optical Character Recognition to Recognise ID cards in the Online Process of Onboarding

Using Image Processing and Optical Character Recognition to ...

引用

2022 International Conference on Smart information systems and Technologies, SIST 2022

作者： Gulnara, Bektemyssova Yerassyl, Akhmer International Information Technology University Almaty Kazakhstan

ISBN: (数字)9781665467902

ISBN: (纸本)9781665467902

A broad range of computer vision studies have been conducted on the recognition of identity documents using mobile devices. A portfolio of techniques and algorithms for solving problems like face recognition, document detection and correction, text field recognition, and so on has been developed, and the lack of data sets has become a critical challenge. The electronic ID card has been commonly used to verify or recognize an individual since its discovery in 2011. Take note of several issues, such as high requirements for detecting the id-card field and challenges recognizing the data in the id-card. In this research, we realized an approach for identifying electronic ID cards using image detection and optical Character Recognition (OCR). This paper provides a short outline of how banking institutions and other fintech companies utilize highly developed OCR tools to automate their processes, enhancing not just service quality, as well as the precision, speed, and protection of their transaction data. As an outcome of our image processing and OCR techniques, we can achieve 99 percent ID detection. This research was based on the interface of a mobile application, such as those owned by one of Kazakhstan's banks. © 2022 IEEE.

关键词： optical character recognition

来源：评论

学校读者我要写书评

暂无评论

Recognition and Analysis of Glass Subsurface Defects by Integrating Machine Vision and Image processing Technology

Recognition and Analysis of Glass Subsurface Defects by Inte...

引用

Computers, information processing and Advanced Education (CIPAE), 2020 International Conference on

作者： Chen Lin Xiong Ling Guangming Luo Wuhan University of Science and Technology Wuhan China Changli Glass Honghu Co. Ltd Wuhan China

ISBN: (数字)9798331527662

ISBN: (纸本)9798331527679

Machine vision and computer image processing technologies are widely used in the metallurgical industry, especially in recognizing and analyzing defects in glass. High surfaces of planer surface and quality in the glass and metal manufacturing sector primarily require automated and performance visual detection, and inspection systems and algorithms are constantly improving. This article, therefore, seeks to comprehensively recognize and analyze glass subsurface defects through the integration of image processing technologies and machine vision. According to the algorithm's properties and the image features typified in the process, the defects of identified glass components are primarily classified based on determining the cause and the optical features of characteristics exhibited. Given the nature of varied and detected effects determined mainly by the classifiers, an improved accuracy level was notable, and a reduction of the computational complexity was evident following the use of machine vision and image processing technologies.

关键词： Visualization Image recognition Machine vision Metals Glass optical imaging Classification algorithms Manufacturing Surface treatment optical devices

来源：评论

学校读者我要写书评

暂无评论

Optimized Approach for Image Design processing in optical Networks: A Comparative Study

引用

Applied Mathematics and Nonlinear Sciences 2024年第1期9卷

作者： Hu, Yunjie Yang, Fan School of Electrical and Information Engineering Wuhan Institute of Technology Hubei Wuhan430000 China

Deep space detection and remote sensing both require optical imaging devices. The optical imaging system often needs a bigger aperture mirror to attain high spatial resolution. As a result, several novel optical imaging systems, such as big segmented mirror telescopes, large aperture membrane diffractive optical telescopes, and others, have been researched in recent years. Real-time wavefront measurement is not required for the wavefront sensorless (WFSless) applied optics (AO) approach. The wavefront corrector is directly regulated via feedback following an image quality measure of the far-field image to correct for wavefront aberration. Integrating artificial neural networks (ANN) and deep learning plays a vital role in developing WFSless AO systems. This paper evaluated various important aspects to provide an in-depth review of the state-of-the-art machine learning-based algorithms deployed in WFSless AO systems. Finally, the applications and prospects were outlined. © 2023 Yunjie Hu et al., published by Sciendo.

关键词： 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：