检索结果-内蒙古大学图书馆

Exploring video Denoising in Thermal Infrared Imaging: Physics-Inspired Noise Generator, Dataset, and Model

IEEE TRANSACTIONS ON image processing 2024年 33卷 3839-3854页

作者： Cai, Lijing Dong, Xiangyu Zhou, Kailai Cao, Xun Nanjing Univ Sch Elect Sci & Engn Nanjing 210023 Peoples R China

We endeavor on a rarely explored task named thermal infrared video denoising. Perception in the thermal infrared significantly enhances the capabilities of machine vision. Nonetheless, noise in imaging systems is one of the factors that hampers the large-scale application of equipment. Existing thermal infrared denoising methods, primarily focusing on the image level, inadequately utilize time-domain information and insufficiently conduct investigation of system-level mixed noise, presenting the inferior ability in the video-recorded era;while video denoising methods, commonly applied to RGB cameras, exhibit uncertain effectiveness owing to substantial dissimilarities in the noise models and modalities between RGB and thermal infrared images. In sight of this, we initially revisit the imaging mechanism, while concurrently introducing a physics-inspired noise generator based on the sources and characteristics of system noise. Subsequently, a thermal infrared video denoising dataset consisting of 518 real-world videos is constructed. Lastly, we propose a denoising model called multi-domain infrared video denoising network, capable of concentrating features from the time, space, and frequency domains to restore high-fidelity videos. Extensive experiments demonstrate that the proposed method achieves state-of-the-art denoising quality and can be successfully applied to commercial cameras and downstream vision tasks, providing a new avenue for clear videography in the thermal infrared world. The dataset and code will be available.

关键词： Thermal infrared video denoising physics-inspired noise generator thermal infrared video denoising dataset multi-domain infrared video denoising network

来源：评论

学校读者我要写书评

暂无评论

real-time Multiclass Face Spoofing Recognition Through Spatiotemporal Convolutional 3D Features 22nd

Real-Time Multiclass Face Spoofing Recognition Through Spati...

引用

22nd International conference on image Analysis and processing (ICIAP)

作者： Giurato, Salvatore Ortis, Alessandro Battiato, Sebastiano Univ Catania Image Proc Lab Dipartimento Matemat & Informat Viale A Doria 6 I-95125 Catania Italy

ISBN: (纸本)9783031510229;9783031510236

Face recognition is used in numerous authentication applications, unfortunately they are susceptible to spoofing attacks such as paper and screen attacks. In this paper, we propose a method that is able to recognise if a face detected in a video is not real and the type of attack performed on the fake video. We propose to learn the temporal features exploiting a 3D Convolution Network that is more suitable for temporal information. The 3D ConvNet, other than summarizing temporal information, allows us to build a real-time method since it is so much more efficient to analyse clips instead of analyzing single frames. The learned features are classified using a binary classifier to distinguish if the person in the clip video is real (i.e. live) or not, multi class classifier recognises if the person is real or the type of attack (screen, paper, ect.). We performed our test on 5 public datasets: Replay Attack, Replay Mobile, MSU-MSFD, Rose-Youtu, RECOD-MPAD.

关键词： Antispoofing Attack 3D Features Multi-Class detection liveness

来源：评论

学校读者我要写书评

暂无评论

Eye-Contact realization Techniques for Improved real time Multimedia Experiences 9

Eye-Contact Realization Techniques for Improved Real Time Mu...

引用

9th International conference on Computer and Communication Systems (ICCCS)

作者： Shi, Junfeng Wang, Guoying Wu, Youlan Cai, Zhifang Hu, Dongqin Beijing Polytech Sch Integrated Circuits Sch Artif Intelligence Beijing Peoples R China

ISBN: (纸本)9798350350227;9798350350210

Systems and methods of multimedia video processing that employ eye-contact realization techniques for improved real time multimedia experiences. Such systems and methods of multimedia video processing are operative to employ multi-view video processing techniques by extracting eye contacts from input multi-view videos and composing input multi-view videos into one single video with free viewpoint eye contents. Having performed such video processing techniques of the input multi-view videos, a high-quality video with free viewpoint eye contacts will be delivered to an end user. By utilizing these novel systems and methods, real time multimedia experiences are significantly improved.

关键词： eye-contact multimedia video multi-view video

来源：评论

学校读者我要写书评

暂无评论

PDSE-YOLOv8: a lightweight detection method for internal defects in asphalt roads

引用

SIGNAL image AND video processing 2024年第12期18卷 8925-8936页

作者： Li, Ning Zhang, Wenliang Liu, Zhaoxu Liu, Kaifeng Wang, Junjie Zhang, Fan Ocean Univ China Dept Civil Engn Qingdao 266100 Peoples R China Shandong Hispeed Grp Co Ltd Innovat Res Inst Jinan 250014 Peoples R China Shandong Hispeed Engn Test CO LTD Jinan 250014 Peoples R China BIM Res Inst China Construct Engn Div 8 Shanghai 200112 Peoples R China

Ground penetrating radar (GPR) is an effective tool for detecting internal defects in asphalt roads due to its non-destructive nature and high resolution. However, detecting defects in GPR images remains challenging, as existing models lack sufficient accuracy and are often complex and redundant. To address these issues, a lightweight real-time detection method based on ground-penetrating radar is proposed in this study. First, a field GPR image dataset of asphalt roads was collected and constructed. To address the limited defect sample data acquired by GPR, an efficient copy-and-paste augmentation method was employed. This method involved copying and pasting defect samples in GPR images while incorporating random scale jitter and position migration operations to generate a sufficient number of real defect samples. Second, the C2f-DSConv module and the SE attention mechanism were designed and introduced based on the YOLOv8 network to improve detection accuracy in the complex background environment of GPR images. Finally, a channel pruning strategy was used to prune the improved YOLOv8 network, reducing model complexity while maintaining detection accuracy. The final model achieves an average detection accuracy of 90.9% and a detection speed of 140.9 FPS. The results show that the proposed method combines both detection accuracy and real-time performance, further advancing the engineering application of internal defect detection in asphalt roads.

关键词： image processing Deep learning algorithm Asphalt pavement Internal defect Channel pruning Ground penetrating radar (GPR)

来源：评论

学校读者我要写书评

暂无评论

LCDnet: a lightweight crowd density estimation model for real-time video surveillance

引用

JOURNAL OF real-time image processing 2023年第2期20卷 29页

作者： Khan, Muhammad Asif Menouar, Hamid Hamila, Ridha Qatar Univ Qatar Mobil Innovat Ctr QM Doha Qatar Qatar Univ Elect Engn Doha Qatar

Automatic crowd counting using density estimation has gained significant attention in computer vision research. As a result, a large number of crowd counting and density estimation models using convolution neural networks (CNN) have been published in the last few years. These models have achieved good accuracy over benchmark datasets. However, attempts to improve the accuracy often lead to higher complexity in these models. In real-time video surveillance applications using drones with limited computing resources, deep models incur intolerable higher inference delay. In this paper, we propose (i) a Lightweight Crowd Density estimation model (LCDnet) for real-time video surveillance, and (ii) an improved training method using curriculum learning (CL). LCDnet is trained using CL and evaluated over two benchmark datasets i.e., DroneRGBT and CARPK. Results are compared with existing crowd models. Our evaluation shows that the LCDnet achieves a reasonably good accuracy while significantly reducing the inference time and memory requirement and thus can be deployed over edge devices with very limited computing resources.

关键词： Crowd counting CNN Density estimation Lightweight real-time

来源：评论

学校读者我要写书评

暂无评论

An integrated and real-time social distancing, mask detection, and facial temperature video measurement system for pandemic monitoring

引用

JOURNAL OF real-time image processing 2023年第5期20卷 95页

作者： Elhanashi, Abdussalam Saponara, Sergio Dini, Pierpaolo Zheng, Qinghe Morita, Daiki Raytchev, Bisser Univ Pisa Ingn Informaz Pisa Italy Shandong Management Univ Sch Intelligent Engn Jinan 250357 Shandong Peoples R China Hiroshima Univ Dept Informat Engn Hiroshima Japan

This paper presents a new Edge-AI algorithm for real-time and multi-feature (social distancing, mask detection, and facial temperature) measurement to minimize the spread of COVID-19 among individuals. COVID-19 has extenuated the need for an intelligent surveillance video system that can monitor the status of social distancing, mask detection, and measure the temperature of faces simultaneously using deep learning (DL) models. In this research, we utilized the fusion of three different YOLOv4-tiny object detectors for each task of the integrated system. This DL model is used for object detection and targeted for real-time applications. The proposed models have been trained for different data sets, which include people detection, mask detection, and facial detection for measuring the temperature, and evaluated on these existing data sets. Thermal and visible cameras have been used for the proposed approach. The thermal camera is used for social distancing and facial temperature measurement, while a visible camera is used for mask detection. The proposed method has been executed on NVIDIA platforms to assess algorithmic performance. For evaluation of the trained models, accuracy, recall, and precision have been measured. We obtained promising results for real-time detection for human recognition. Different couples of thermal and visible cameras and different NVIDIA edge platforms have been adopted to explore solutions with different trade-offs between cost and performance. The multi-feature algorithm is designed to monitor the individuals continuously in the targeted environments, thus reducing the impact of COVID-19 spread.

关键词： Multi-feature detection Social distancing Mask detection Facial temperature real-time COVID-19 YOLOv4-tiny Edge AI video platforms

来源：评论

学校读者我要写书评

暂无评论

DGC-Net: Dynamic Graph Contrastive Network for video Object Detection

引用

IEEE TRANSACTIONS ON image processing 2025年 34卷 2269-2284页

作者： Qi, Qiang Wang, Hanzi Yan, Yan Li, Xuelong Qingdao Univ Sci & Technol Sch Data Sci Qingdao 266061 Peoples R China Xiamen Univ Sch Informat Fujian Key Lab Sensing & Comp Smart City Xiamen 361005 Peoples R China China Telecom Inst Artificial Intelligence TeleAI Beijing 100033 Peoples R China China Telecom Inst Artificial Intelligence TeleAI Shanghai 200030 Peoples R China

video object detection is a challenging task in computer vision since it needs to handle the object appearance degradation problem that seldom occurs in the image domain. Off-the-shelf video object detection methods typically aggregate multi-frame features at one stroke to alleviate appearance degradation. However, these existing methods do not take supervision knowledge into consideration and thus still suffer from insufficient feature aggregation, resulting in the false detection problem. In this paper, we take a different perspective on feature aggregation, and propose a dynamic graph contrastive network (DGC-Net) for video object detection, including three improvements against existing methods. First, we design a frame-level graph contrastive module to aggregate frame features, enabling our DGC-Net to fully exploit discriminative contextual feature representations to facilitate video object detection. Second, we develop a proposal-level graph contrastive module to aggregate proposal features, making our DGC-Net sufficiently learn discriminative semantic feature representations. Third, we present a graph transformer to dynamically adjust the graph structure by pruning the useless nodes and edges, which contributes to improving accuracy and efficiency as it can eliminate the geometric-semantic ambiguity and reduce the graph scale. Furthermore, inherited from the framework of DGC-Net, we develop DGC-Net Lite to perform real-time video object detection with a much faster inference speed. Extensive experiments conducted on the imageNet VID dataset demonstrate that our DGC-Net outperforms the performance of current state-of-the-art methods. Notably, our DGC-Net obtains 86.3%/87.3% mAP when using ResNet-101/ResNeXt-101.

关键词： Feature extraction Object detection Streaming media Aggregates Transformers Contrastive learning Accuracy Proposals real-time systems image edge detection video object detection graph contrastive learning graph transformer feature aggregation

来源：评论

学校读者我要写书评

暂无评论

The Art System of Picture Composition Based on Computer video processing Technology 2

The Art System of Picture Composition Based on Computer Vide...

引用

2nd IEEE International conference on Distributed Computing and Electrical Circuits and Electronics, ICDCECE 2023

作者： Qiu, Shan Wang, Ning Zibo Technician College Shandong China

ISBN: (纸本)9798350347456

With the development of science and technology and the renewal of media means in the era of cultural industry, electronic media develops rapidly, video image technology with digital realization as the carrier develops rapidly, and gradually becomes an independent art form, which indicates the arrival of a new type of art. The purpose of this paper is to study the picture composition art system based on computer video processing technology. From the picture composition art overview to the introduction of basic knowledge of computer science, and then to the main technology of computer video processing, summed up the coordinate space and its transformation, geometric transformation technology of graphics and other knowledge. This paper mainly analyzes the program flow of picture composition art system and modern geometric composition law module, and introduces the application of video frame in image processing. The real-time performance and effect of the system are tested. When processing video art image composition, the CPU usage of real-time processing is about 40%, If you don't process the video, but just play it, the average CPU usage is about 20%. The real-time performance of video processing can be realized effectively through the image synthesis art system. © 2023 IEEE.

关键词： video signal processing

来源：评论

学校读者我要写书评

暂无评论

PCB defect detection based on PSO-optimized threshold segmentation and SURF features

引用

SIGNAL image AND video processing 2024年第5期18卷 4327-4336页

作者： Chang, Yuanpei Xue, Ying Zhang, Yu Sun, Jingguo Ji, Zhangyuan Li, Hewei Wang, Teng Zuo, Jiancun Shanghai Polytech Univ Sch Intelligent Mfg & Control Engn Shanghai 201209 Peoples R China Shanghai Polytech Univ Inst Artificial Intelligence Sch Comp & Informat Engn Shanghai 201209 Peoples R China

This paper proposes an automated inspection approach for printed circuit boards (PCBs) that can accurately locate defects to solve the issues of low precision, complex equipment, and high cost. Digital image processing techniques are utilized in this method, including filtering, image segmentation, feature extraction, alignment, and mathematical morphology processing. To overcome the Otsu thresholding segmentation algorithm's high computational cost and poor real-time performance, a particle swarm approach is optimized to increase image segmentation efficiency. Meanwhile, combining the benefits of the FLANN algorithm and the SURF method, matching image feature points is done based on the SURF algorithm. The performance of matching image feature points is improved. In addition, the alignment error of the images is reduced. According to experimental results, the improved PCB defect detection algorithm demonstrated 98.9% accuracy, with remarkable efficiency and accuracy, and can satisfy PCB defect detection requirements.

关键词： image processing Printed circuit board SURF PSO Defect detection

来源：评论

学校读者我要写书评

暂无评论

Enhancement of Coastline video Monitoring System Using Structuring Element Morphological Operations

引用

INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS 2025年第1期16卷 603-611页

作者： Pawana, I. Gusti Ngurah Agung Widyantara, I. Made Oka Sudarma, Made Wiharta, Dewa Made Jayantari, Made Widya Udayana Univ Elect Engn Dept Badung Indonesia Udayana Univ Civil Engn Dept Badung Indonesia

Coastal monitoring is vital in environmental management, disaster mitigation, and addressing climate change impacts. Traditional methods are time-consuming and error- prone, prompting the need for innovative systems. This study introduces the Coastal video Monitoring System (CoViMos), a novel framework for real-time shoreline detection in tropical regions, specifically at Kedonganan Beach, Bali. The CoViMos framework utilizes advanced video monitoring and optimized morphological operations to address challenges such as environmental noise and dynamic shoreline behavior. Key innovations include Kapur's entropy thresholding enhanced with the Grasshopper Optimization Algorithm (GOA) and structuring elements tailored to the beach's unique features. Sensitivity analysis reveals that a structuring element size of five pixels offers optimal performance, balancing efficiency, and image fidelity. This configuration achieves peak values in quality metrics such as the Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), Complex Wavelet SSIM (CWSSIM), and Feature Similarity Index (FSIM) while minimizing Mean Squared Error (MSE) and reducing processing time. The results demonstrate significant improvements in shoreline detection accuracy, with PSNR increasing by 9.3%, SSIM by 1.4%, CWSSIM by 1.7%, and FSIM by 1.6%. processing time decreased by 1.3%, emphasizing the system's computational efficiency. These enhancements ensure more precise shoreline mapping, even in noisy and dynamic environments.

关键词： Coastline detection image processing video Monitoring System (CoViMoS) morphological operations

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：