检索结果-内蒙古大学图书馆

2nd International conference on Image Processing, computer vision and Machine Learning, ICICML 2023

作者： Li, Hong Hussin, Norriza Pingdingshan University College of Information Engineering Pingdingshan467002 China SEGi University Faculty of Engineering Built Environment and Information Technology Kota Damansara47810 Malaysia

ISBN: (纸本)9798350331417

The conventional visual identification method of road sign line loss mainly uses Sobel\Prewitt operator to dynamically divide the visual identification area, which is easily affected by the directional change of road signs, resulting in poor visual identification performance. Therefore, it is necessary to design a brand-new visual identification method of road sign line loss based on image high-resolution reconstruction algorithm. That is to say, the visual recognition features of road sign line missing are extracted, and the visual recognition algorithm of road sign line image super-resolution reconstruction is designed, thus realizing the visual recognition of road sign line missing. The experimental results show that the designed visual recognition method for high-resolution reconstruction of road marking line images has good recognition performance, accuracy and certain application value, and has made certain contributions to improving the safety of road traffic. © 2023 ieee.

关键词： Identification Identification line Image high resolution reconstruction algorithm Missing Road vision Way

来源：评论

学校读者我要写书评

暂无评论

Automated Radar Data Labeling using MoveNet for Human Gesture recognition 31

Automated Radar Data Labeling using MoveNet for Human Gestur...

引用

31st Asia-Pacific Microwave conference (APMC) - Microwave Linking the World

作者： Maiwald, Timo Gabsteiger, Jasmin Weigel, Robert Lurz, Fabian Friedrich Alexander Univ Inst Elect Engn Erlangen Germany Hamburg Univ Technol Inst High Frequency Technol Hamburg Germany

ISBN: (纸本)9781665494182

Integrated radar sensors constitute an emerging technology used from smart home to autonomous driving applications. In combination with machine learning for radar data signal processing, they experience increasing popularity and are heavily investigated. These methods however need huge amounts of labeled training data to perform well in realistic scenarios, which is time consuming and cost intensive. To increase data acquisition speed, this paper presents a system using a fast computer vision model for automated radar data labeling. It is capable of generating 10 labeled radar data frames per second. To demonstrate the abilities it is used for human gesture recognition.

关键词： gesture recognition fmcw radar computer vision machine learning

来源：评论

学校读者我要写书评

暂无评论

Real-time Bangla License Plate recognition System for Low Resource Video-based Applications

Real-time Bangla License Plate Recognition System for Low Re...

引用

22nd ieee/CVF Winter conference on Applications of computer vision (WACV)

作者： Ashrafee, Alif Khan, Akib Mohammed Irbaz, Mohammad Sabik Al Nasim, Md Abdullah Islamic Univ Technol Dept Comp Sci & Engn Gazipur Bangladesh Pioneer Alpha Ltd Machine Learning Team Dhaka Bangladesh

ISBN: (纸本)9781665458245

Automatic License Plate recognition systems aim to provide a solution for detecting, localizing, and recognizing license plate characters from vehicles appearing in video frames. However, deploying such systems in the real world requires real-time performance in low-resource environments. In our paper, we propose a two-stage detection pipeline paired with vision API that provides real-time inference speed along with consistently accurate detection and recognition performance. We used a haar-cascade classifier as a filter on top of our backbone MobileNet SSDv2 detection model. This reduces inference time by only focusing on high confidence detections and using them for recognition. We also impose a temporal frame separation strategy to distinguish between multiple vehicle license plates in the same clip. Furthermore, there are no publicly available Bangla license plate datasets, for which we created an image dataset and a video dataset containing license plates in the wild. We trained our models on the image dataset and achieved an AP(0.5) score of 86% and tested our pipeline on the video dataset and observed reasonable detection and recognition performance (82.7% detection rate, and 60.8% OCR F1 score) with real-time processing speed (27.2 frames per second).

关键词： computer vision Image recognition conferences Pipelines Focusing Licenses Streaming media

来源：评论

学校读者我要写书评

暂无评论

Semi-automatic 3D Object Keypoint Annotation and Detection for the Masses 26

Semi-automatic 3D Object Keypoint Annotation and Detection f...

引用

26th International conference on pattern recognition / 8th International Workshop on Image Mining - Theory and Applications (IMTA)

作者： Blomqvist, Kenneth Chung, Jen Jen Ott, Lionel Siegwart, Roland Swiss Fed Inst Technol Autonomous Syst Lab Zurich Switzerland

ISBN: (数字)9781665490627

ISBN: (纸本)9781665490627

Creating computer vision datasets requires careful planning and lots of time and effort. In robotics research, we often have to use standardized objects, such as the YCB object set, for tasks such as object tracking, pose estimation, grasping and manipulation, as there are datasets and pre-learned methods available for these objects. This limits the impact of our research since learning-based computer vision methods can only be used in scenarios that are supported by existing datasets. In this work, we present a full object keypoint tracking toolkit, encompassing the entire process from data collection, labeling, model learning and evaluation. We present a semi-automatic way of collecting and labeling datasets using a wrist mounted camera on a standard robotic arm. Using our toolkit and method, we are able to obtain a working 3D object keypoint detector and go through the whole process of data collection, annotation and learning in just a couple hours of active time.

关键词： Wrist computer vision Three-dimensional displays Annotations Robot vision systems Pose estimation Data collection

来源：评论

学校读者我要写书评

暂无评论

FishEye8K: A Benchmark and Dataset for Fisheye Camera Object Detection

FishEye8K: A Benchmark and Dataset for Fisheye Camera Object...

引用

2023 ieee/CVF conference on computer vision and pattern recognition Workshops, CVPRW 2023

作者： Gochoo, Munkhjargal Otgonbold, Munkh-Erdene Ganbold, Erkhembayar Hsieh, Jun-Wei Chang, Ming-Ching Chen, Ping-Yang Dorj, Byambaa Al Jassmi, Hamad Batnasan, Ganzorig Alnajjar, Fady Abduljabbar, Mohammed Lin, Fang-Pang United Arab Emirates University Department of Computer Science and Software Engineering United Arab Emirates United Arab Emirates University Emirates Center for Mobility Research United Arab Emirates National Yang Ming Chiao Tung University College of AI and Green Energy Taiwan University at Albany - State University of New York NY United States National Yang Ming Chiao Tung University Department of Computer Science Taiwan Mongolian University of Science and Technology Mongolia National Center for High-Performance Computing Taiwan

ISBN: (纸本)9798350302493

With the advance of AI, road object detection has been a prominent topic in computer vision, mostly using perspective cameras. Fisheye lens provides omnidirectional wide coverage for using fewer cameras to monitor road intersections, however with view distortions. To our knowledge, there is no existing open dataset prepared for traffic surveillance on fisheye cameras. This paper introduces an open FishEye8K benchmark dataset for road object detection tasks, which comprises 157K bounding boxes across five classes (Pedestrian, Bike, Car, Bus, and Truck). In addition, we present benchmark results of State-of-The-Art (SoTA) models, including variations of YOLOv5, YOLOR, YOLO7, and YOLOv8. The dataset comprises 8,000 images recorded in 22 videos using 18 fisheye cameras for traffic monitoring in Hsinchu, Taiwan, at resolutions of 1080×1080 and 1280×1280. The data annotation and validation process were arduous and time-consuming, due to the ultra-wide panoramic and hemispherical fisheye camera images with large distortion and numerous road participants, particularly people riding scooters. To avoid bias, frames from a particular camera were assigned to either the training or test sets, maintaining a ratio of about 70:30 for both the number of images and bounding boxes in each class. Experimental results show that YOLOv8 and YOLOR outperform on input sizes 640×640 and 1280×1280, respectively. The dataset will be available on the GitHub link with PASCAL VOC, MS COCO, and YOLO annotation formats. The FishEye8K benchmark will provide significant contributions to the fisheye video analytics and smart city applications. © 2023 ieee.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Design and Implementation of a vision- and Grating-Sensor-Based Intelligent Unmanned Settlement System

IEEE Transactions on Artificial Intelligence

引用

ieee Transactions on Artificial Intelligence 2022年第2期3卷 254-264页

作者： Zhang, Hong-Bo Zhou, Yi-Zhong Dong, Li-Jia Lei, Qing Du, Ji-Xiang Department of Computer Science and Technology Huaqiao University Xiamen361000 China Xiamen Key Laboratory of Computer Vision and Pattern Recognition The Fujian Key Laboratory of Big Data Intelligence and Security Huaqiao University Xiamen361000 China

In this article, a new vision- and grating-sensor-based intelligent unmanned settlement (IUS) system is proposed for convenience stores to automatically recognize the shopping behavior of customers, record their identities, and generate invoices. First, we design a new IUS architecture, which includes a shelf module and exit module. To achieve automatic settlement for each customer, a shopping event detection method is proposed. In this method, a vision-based human pose estimation algorithm is used to detect a human form standing in front of a shelf. The hand actions of each customer are detected by a grating sensor, and an image recognition method based on a convolutional neural network (CNN) is applied to recognize the items in the hands of customers. To reduce the image annotation workload, we propose a semisupervised training method for the recognition network. Based on hand action detection and item recognition, a shopping event recognition method is designed for the system, and a facial image of the customer corresponding to each shopping behavior is captured. Finally, each detected shopping event is added to the invoice of the corresponding customer via a facial recognition method. To verify the effectiveness of the proposed IUS system, we have built a handheld item image dataset and a shopping event dataset for an unmanned convenience store. The experimental results show that the proposed system can accurately recognize shopping behaviors and generate invoices. © 2020 ieee.

关键词： Sales

来源：评论

学校读者我要写书评

暂无评论

Unveiling the vision: A Comprehensive Review of computer vision in AI and ML

Unveiling the Vision: A Comprehensive Review of Computer Vis...

引用

2024 International conference on Advances in Data Engineering and Intelligent Computing Systems, ADICS 2024

作者： Laad, Meena Maurya, Ratan Saiyed, Najeeb Symbiosis Institute Of Technology Dept. Of Applied Science Maharashtra Pune India Symbiosis Institute Of Technology Dept. Of Artificial Intelligence And Machine Learning Maharashtra Pune India

ISBN: (纸本)9798350364828

In the era of rapid technological advancement, computer vision has emerged as a transformative force, reshaping the landscape of Artificial Intelligence (AI) and Machine Learning (ML). This comprehensive review paper aims to delve into the intricate evolution, methodologies, applications, challenges, and future trajectories of computer vision. Moving beyond a mere exploration of technical intricacies, our objective is to present a holistic narrative that encapsulates the profound impact of computer vision on AI and ML and its repercussions across society. The journey begins by traversing the philosophical and historical roots of computer vision, unraveling the threads that connect human visual perception to the development of artificial vision. By exploring the historical evolution from early image processing to the current era of deep learning, we seek to elucidate the intellectual milestones that have shaped the field. Methodologically, this paper navigates through both traditional approaches and contemporary deep learning paradigms. It dissects traditional methods, emphasizing their enduring relevance and influence on modern computer vision applications. In parallel, exploring deep learning delves into established architectures all the nuanced impact of design choices on interpretability and explain ability. Applications form a cornerstone of our review, with an enriched focus on case studies that spotlight the transformative influence of computer vision. Beyond the traditional domains of image recognition, we delve into the healthcare renaissance, where computer vision contributes to diagnostics, drug discovery, and personalized medicine. Furthermore, we explore its role in smart cities, extending beyond surveillance to urban planning, traffic management, and environmental monitoring. © 2024 ieee.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

Finger Pointnet: An End-to-end Embedded Network for 3D Fingerprint Verification 4

Finger Pointnet: An End-to-end Embedded Network for 3D Finge...

引用

4th ieee International conference on pattern recognition and Machine Learning, PRML 2023

作者： Huang, Yixin Liu, Feng Yang, Dongliang Song, Changjiang Shenzhen University Shenzhen China Heilongjiang Academy of Sciences Intelligent Manufacturing Research Institute Heilongjiang China

ISBN: (纸本)9798350324303

Contactless 3D fingerprint gains significant attention in recent years. Much of previous works for 3D fingerprint verification were based on minutiae. They required a series of laborious preprocessing and relied on software to extract minutiae. To alleviate these problems, this research presents an end-to-end embedded network for fingerprint verification. Firstly, we propose to extract ridge lines from the original large and redundant 3D fingerprint point cloud data. Then, we design two weight-shared encoders to extract features from a pair of positive 3D fingerprints. Finally, loss backpropagation from the hardest sample in a mini-batch permitted in our approach makes it can distinguish negative samples. The experiments on a public database and an in-house database verify that our approach achieves desirable performance and significantly speeds up feature extraction and matching compared with traditional approaches. © 2023 ieee.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

HyperStyle: StyleGAN Inversion with HyperNetworks for Real Image Editing

HyperStyle: StyleGAN Inversion with HyperNetworks for Real I...

引用

2022 ieee/CVF conference on computer vision and pattern recognition, CVPR 2022

作者： Alaluf, Yuval Tov, Omer Mokady, Ron Gal, Rinon Bermano, Amit Blavatnik School of Computer Science Tel Aviv University Israel

ISBN: (数字)9781665469463

ISBN: (纸本)9781665469463

The inversion of real images into StyleGAN's latent space is a well-studied problem. Nevertheless, applying existing approaches to real-world scenarios remains an open challenge, due to an inherent trade-off between reconstruction and editability: latent space regions which can accurately represent real images typically suffer from degraded semantic control. Recent work proposes to mitigate this trade-off by fine-tuning the generator to add the target image to well-behaved, editable regions of the latent space. While promising, this fine-tuning scheme is impractical for prevalent use as it requires a lengthy training phase for each new image. In this work, we introduce this approach into the realm of encoder-based inversion. We propose HyperStyle, a hypernetwork that learns to modulate StyleGAN's weights to faithfully express a given image in editable regions of the latent space. A naive modulation approach would require training a hypernetwork with over three billion parameters. Through careful network design, we reduce this to be in line with existing encoders. HyperStyle yields reconstructions comparable to those of optimization techniques with the near real-time inference capabilities of encoders. Lastly, we demonstrate HyperStyle's effectiveness on several applications beyond the inversion task, including the editing of out-of-domain images which were never seen during training. Code is available on our project page: https://***/hyperstyle/. © 2022 ieee.

关键词： Training Semantics Aerospace electronics Generators Robustness Real-time systems pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Based pattern recognition with HOG Features for Detection of Islanding 3

Machine Learning Based Pattern Recognition with HOG Features...

引用

3rd ieee International conference on Electrical Power and Energy Systems, ICEPES 2024

作者： Pal, Kumaresh Goswami, Subhadip Bhuyan, Anshuman Koley, Santanu Bera, Bidisha Basu, Sarmistha National Institute of Technology Department of Electrical Engineering Jamshedpur India National Institute of Technology Department of Electrical Engineering Arunachal Pradesh India Gandhi Institute for Education & Technology Department of Electrical Engineering Bhubaneswar India Haldia Institute of Technology Department of Computer Science Engineering Haldia India Global Institute of Science and Technology Department of Computer Science and Technology Haldia India Calcutta Institute of Engineering and Management Department of Physics Kolkata India

ISBN: (纸本)9798350390728

Islanding is a significant difficulty that arises from the high integration of distributed generation sources. Islanding could potentially damage the clients and their equipment. According to the ieee 1547 DG interconnection requirements, islanding must be detected within two seconds, and the DG must be shut down. This research introduces a new method for identifying islanding using image pattern recognition utilising a Machine Learning classification algorithm. The suggested technique utilises the Random Forest (RF) classification algorithm. Images are utilised to extract the histogram of oriented gradient (HOG) features in order to detect both non-islanding and islanding occurrences. The spectrogram images are formed from the time-series signal collected from the point of common coupling. The images are utilised to extract features from histogram of oriented gradient, which are subsequently used as feature vector inputs for training and testing the RF classifiers. The spectrogram image is derived using the rate of change of the voltage parameter. The efficacy of the RF classifier is evaluated using 5, 10, 20, and 25-fold cross-validations. The classification findings indicates that islanding detection utilising a Random Forest classifier and Histogram of Oriented Gradients (HOG) feature for image pattern recognition obtained outstanding performance with 98.75% accuracy and a detection time of 192 milliseconds. © 2024 ieee.

关键词： Spectrographs

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：