检索结果-内蒙古大学图书馆

15th International conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Shubha, B. Veena Devi Shastrimath, V. Shreesha, C. Department of E and Ce Nitte574110 India Manipal Institute of Technology Manipal Academy of Higher Education Department of I and Ce Manipal576104 India

ISBN: (纸本)9798350370249

image processing and analysis make extensive use of image scaling. In digital image processing, resizing an image is referred to as image scaling. When images are magnified, one of the most important factors is their resolution. There have been several interpolation techniques proposed to enhance the image quality. This paper's primary objective is to provide a comparative analysis of different interpolation methods like nearest neighbor, bilinear, and bicubic interpolation for thermal images in real-time. It is observed from the result that by using the bicubic interpolation method the average Peak Signal to Noise Ratio (PSNR) is 29dB which is high by 1 dB compared to bilinear and nearest neighbor interpolation methods. The Mean Square Error (MSE) value is 26 which is less by approximately 4 compared to bilinear and nearest neighbor interpolation methods. The research work is carried out by reading the thermal images from the thermopile sensor using Raspberry Pi in real time. © 2024 IEEE.

关键词： Bicubic Bilinear image interpolation Nearest Neighbor Thermopile Sensor

来源：评论

学校读者我要写书评

暂无评论

real-time Obstacle Detection and Path Adjustment for Autonomous Robots Using OpenCV

Real-Time Obstacle Detection and Path Adjustment for Autonom...

引用

2024 International conference on Telecommunications and Intelligent Systems, ICTIS 2024

作者： Elaoua, Abrar Nadour, Mohamed Cherroun, Lakhmissi Mahdi, Mohammed Taqiyeddine Hafaifa, Ahmed Djelfa Algeria

ISBN: (纸本)9798331527396

The real-time obstacle detection and path adjustment system for autonomous robots presented in this paper was created using OpenCV. The combination of image processing techniques enables the robot to identify and navigate around any obstacle efficiently. The main functions include real-time image acquisition;where the robot constantly takes pictures of its surroundings, and obstacle detection;which uses edge detection and contour analysis for proper identification of hazards in the environment. The method entails that once an object is detected, the robot's route is reshaped with sophisticated algorithms that are aimed at optimizing navigation as well as minimizing collision probabilities. A wide range of challenging settings such as static or dynamic environments were subjected to rigorous testing on the proposed system. The results indicate that it is robust and effective in detecting obstacles reliably to enhance navigation based on its capacity to make adjustments during operation time (discovered employing experimentation that took place in these settings). © 2024 IEEE.

关键词： image acquisition

来源：评论

学校读者我要写书评

暂无评论

COVARIANCE-AWARE FEATURE ALIGNMENT WITH PRE-COMPUTED SOURCE STATISTICS FOR TEST-time ADAPTATION TO MULTIPLE image CORRUPTIONS 30

COVARIANCE-AWARE FEATURE ALIGNMENT WITH PRE-COMPUTED SOURCE ...

引用

30th IEEE International conference on image processing (ICIP)

作者： Adachi, Kazuki Yamaguchi, Shin'ya Kumagai, Atsutoshi NTT Comp & Data Sci Labs Tokyo Japan Kyoto Univ Kyoto Japan

ISBN: (纸本)9781728198354

real-world image recognition systems often face corrupted input images, which cause distribution shifts and degrade the performance of models. These systems often use a single prediction model in a central server and process images sent from various environments, such as cameras distributed in cities or cars. Such single models face images corrupted in heterogeneous ways in test time. Thus, they require to instantly adapt to the multiple corruptions during testing rather than being re-trained at a high cost. Test-time adaptation (TTA), which aims to adapt models without accessing the training dataset, is one of the settings that can address this problem. Existing TTA methods indeed work well on a single corruption. However, the adaptation ability is limited when multiple types of corruption occur, which is more realistic. We hypothesize this is because the distribution shift is more complicated, and the adaptation becomes more difficult in case of multiple corruptions. In fact, we experimentally found that a larger distribution gap remains after TTA. To address the distribution gap during testing, we propose a novel TTA method named Covariance-Aware Feature alignment (CAFe). We empirically show that CAFe outperforms prior TTA methods on image corruptions, including multiple types of corruptions.

关键词： Neural network test-time adaptation image corruption

来源：评论

学校读者我要写书评

暂无评论

Application Research of On-Board Satellite Remote Sensing image Terrain Classification Based on Deep Learning 11th

Application Research of On-Board Satellite Remote Sensing Im...

引用

11th International conference on Signal and Information processing, Network and Computers, ICSINC 2023

作者： Tian, Hexiang Zhang, Jingya Zhao, Wenrui Institute of Remote Sensing Satellite CAST Beijing100094 China

ISBN: (纸本)9789819721238

Aiming at the real-time on-board intelligent decoding/translating need of the satellite remote sensing image, the article raises up a system structure of satellite remote sensing image terrain classification based on the combination of terrain servers and on-board AI analyzing platform, and designs an on-board satellite remote sensing image terrain classification algorithm based on deepabv3+. Then, the optimizing solution satisfies the constraints of the performance and the storage of the on-board intelligent information system is offered. Tested by on-board AI analyzing platform built on land, the solution is proved to be able to satisfy the on-board real-time processing need, the on-board image decoding/translating speed and accuracy requirements. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Satellites

来源：评论

学校读者我要写书评

暂无评论

Neuromorphic Chiplet Architecture for Wide Area Motion imagery processing 6

Neuromorphic Chiplet Architecture for Wide Area Motion Image...

引用

6th Argentine conference on Electronics (CAE)

作者： Andreou, Andreas G. Figliolia, Tomas Sanni, Kayode Murray, Thomas S. Tognetti, Gaspar Mendat, Daniel R. Molin, Jamal L. Villemur, Martin Pouliquen, Philippe O. Julian, Pedro Etienne-Cummings, Ralph Doxas, Isidoros Johns Hopkins Univ Dept Elect & Comp Engn Baltimore MD 21218 USA BAE Syst London England Johns Hopkins Univ Baltimore MD 21218 USA

ISBN: (纸本)9798350305081

We present the system architecture for real-time processing of data that originates in large format tiled imaging arrays used in wide area motion imagery ubiquitous surveillance. High performance and high throughput is achieved through approximate computing and fixed point variable precision (6 bits to 18 bits) arithmetic. The architecture implements a variety of processing algorithms in what we consider today as Third Wave AI and Machine Intelligence ranging from convolutional networks (CNNs) to linear and non-linear morphological processing, probabilistic inference using exact and approximate Bayesian methods and Deep Neural Networks based classification. The processing pipeline is implemented entirely using event based neuromorphic and stochastic computational primitives. An emulation of the system architecture demonstrated processing in real-time 160 x 120 raw pixel data running on a reconfigurable computing platform (5 Xilinx Kintex-7 FPGAs). The reconfigurable computing implementation was developed to emulate the computational structures for a 2.5D System chiplet design, that was fabricated in the 55nm GF CMOS technology. To optimize for energy efficiency of a mixed level system, a general energy aware methodology is applied through the design process at all levels from algorithms and architecture all the way down to technology and devices, while at the same time keeping the operational requirements and specifications for the task at focus.

关键词： Chiplets 2.5D architecture neuromorphic processing mixed signal design event based processing wide area image processing

来源：评论

学校读者我要写书评

暂无评论

Multi-DSP Joint SAR real-time Imaging System Based on PFA Sub-Aperture Imaging Algorithm 2

Multi-DSP Joint SAR Real-Time Imaging System Based on PFA Su...

引用

2nd IEEE International conference on Signal, Information and Data processing, ICSIDP 2024

作者： Liu, Xueqing Song, Yue Li, Zhongyu Wu, Junjie Yang, Jianyu Du, Ping Zhang, Yueyue Dong, Mingji Miao, Xin Liu, Jiaxuan University of Electronic Science and Technology of China School of Information and Communication Engineering Chengdu China Shanghai Satellite Network Research Institute Co. Ltd. Shanghai Key Laboratory of Satellite Network Shanghai China

ISBN: (纸本)9798331515669

The Synthetic Aperture Radar(SAR) real-time imaging system designed with multi-DSP architecture has been widely used in both airborne and satellite platforms. In the context of real-time imaging systems, Polar Format Algorithm(PFA) fall short of meeting the system's real-time demands. This paper proposes a multi DSP architecture SAR imaging system based on sub-aperture PFA to meet the real-time imaging requirements. The system utilizes multiple sub-aperture echo data to design a "2+1"DSP C6678 hardware architecture, and implements the sub aperture PFA algorithm in parallel on the hardware to improve imaging efficiency. Furthermore, we performed phase gradient autofocus processing on the imaging results and achieved high-resolution images. The experimental results show that compared with traditional PFA, the proposed sub aperture PFA algorithm and hardware architecture reduce imaging time by 54%. © 2024 IEEE.

关键词： image enhancement

来源：评论

学校读者我要写书评

暂无评论

realtime Scene Enhancement of Low Light image/Video 5

Realtime Scene Enhancement of Low Light Image/Video

引用

5th International conference on Intelligent Computing and Human-Computer Interaction, ICHCI 2024

作者： Zeng, Huixia Xiao, Ping School of Electronic and Information Engineering Guangzhou City University of Technology Guangdong Guangzhou510800 China

ISBN: (纸本)9798350368284

The quality of image and videos plays a vital role in case of real-time systems. images are captured without sufficient illumination, lead to low dynamic range and high propensity for generating high noise levels. The processing ideas and networks of several mainstream low-light image enhancement neural network algorithms such as DRBN, DSLR, ZeroDCE, EnlightenGAN, as well as the key technologies used are compared and analyzed. In this paper, the EnglightenGAN neural network is used as the core image enhancement algorithm, and NVIDIA AGX Xavier equipment and video stream processing technology are used to achieve the entire process of video acquisition, video frame encoding, image enhancement processing, frame decoding, and video stream display. It can adapt to most low-light and dark scenes which has a good enhancement effect on video streams collected in low-light scenes. The real-time performance and image enhancement effect of the system achieved good results, with FPS reaching 10, SSIM reaching 0.51753, and PSNR reaching 27.89973. © 2024 IEEE.

关键词： Video streaming

来源：评论

学校读者我要写书评

暂无评论

real-time target detection method of UAV aerial image combined with X-ray digital imaging technology 5

Real-time target detection method of UAV aerial image combin...

引用

5th International conference on Computer Vision, image and Deep Learning, CVIDL 2024

作者： Chen, Hui Wang, Fangfang Tan, Xinghua Yu, Long Li, Zhenghui Xinjiang Power Transmission and Transformation Co. Ltd Xinjiang Urumqi830000 China Henan Star Detection Technology Co. Ltd Henan Xuchang461000 China

ISBN: (纸本)9798350373820

Due to the influence of weather, illumination and other factors, the image quality of UAV aerial images is not good, so a real-time target detection method of UAV aerial images combined with X-ray digital imaging technology is proposed. Following the attenuation law, the intensity of the target detection and registration search area is described by the quality degradation coefficient, the sampling points of the aerial image of the UAV are configured, and the target detection and registration search area is delineated by combining the X -ray digital imaging technology;On this basis, the high-definition images taken by UAV are analyzed in real time, and the real-time target marking points of UAV aerial images are identified. The X-ray digital imaging technology is applied to the images, and the effective area is extracted and the calculated actual value is output as the output result of real-time target detection of UAV aerial images, and the defogging processing result of UAV aerial images is obtained. The experimental results show that this method improves the definition of aerial images, effectively removes the interference information such as clouds in aerial images, and has high detection accuracy and distortion correction on targets such as trees, roads, houses and farmland, and has good detection performance. © 2024 IEEE.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Applying Deep Learning and Heatmap Techniques for Sinusitis Diagnosis Using Enhanced 3-D image Segmentation

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2025年 74卷

作者： Huang, Pin-Siang Horng, Shi-Jinn Yang, Chia-Jung Chen, Hongmei Zhou, Wanlei Natl Taiwan Univ Sci & Technol Dept Comp Sci & Informat Engn Taipei 10607 Taiwan Asia Univ Dept Comp Sci & Informat Engn Taichung 41354 Taiwan China Med Univ China Med Univ Hosp Dept Med Res Taichung Taiwan Univ Technol Sydney Sch Comp Sci Ultimo NSW 2007 Australia Taitung MacKay Mem Hosp Dept Radiol Taitung 950 Taiwan Southwest Jiaotong Univ Sch Comp & Artificial Intelligence Chengdu 610032 Peoples R China City Univ Macau Fac Data Sci Macau Peoples R China

Sinusitis, characterized by inflammation of the sinus cavities, presents diagnostic challenges due to the subjective and time-consuming nature of manual computed tomography (CT) scan annotations. This study introduces a deep learning (DL) method combined with heatmap analysis to automate 3-D segmentation of sinusitis in CT images. The proposed approach reduces the dependency on fully segmented annotations, instead requiring only binary classification labels during training. A classification model identifies areas likely affected by inflammation, highlighted by gradient-weighted class activation mapping (Grad-CAM) to produce heatmaps, which are refined using a watershed algorithm for precise segmentation. This approach reduces annotation preparation time from 30 min to under 1 min while achieving a Dice similarity coefficient (DSC) of 90.04%. Built on the EfficientNet-B2 architecture, the model outperforms state-of-the-art methods like U-Net, U-Net++, and statistical-based approaches. Additionally, a novel instrumentation system integrated with CT scanners calculates inflammation areas in real time, streamlining diagnostic workflows. This method accelerates the process while improving the consistency and reliability of sinusitis diagnosis, potentially transforming clinical practices with advanced image processing techniques.

关键词： image segmentation Computed tomography Deep learning Watersheds Three-dimensional displays Accuracy Medical diagnostic imaging Convolutional neural networks Annotations Training 3-D segmentation computed tomography (CT) scan deep learning (DL) diagnostic automation Dice similarity coefficient (DSC) gradient-weighted class activation mapping (GRAD-CAM) image annotation medical image processing sinusitis watershed algorithm

来源：评论

学校读者我要写书评

暂无评论

Impacts of image Compression on the Detection Quality of a Novel real-time image processing Platform 26

Impacts of Image Compression on the Detection Quality of a N...

引用

IEEE 26th International conference on Intelligent Transportation Systems (ITSC)

作者： Mehrke, Jannik Volk, Georg Stumpp, Yannik Bringmann, Oliver Terzis, Anestis Ulm Univ Appl Sci Inst Commun Technol Ulm Germany Univ Tubingen Dept Comp Engn Tubingen Germany

ISBN: (纸本)9798350399462

Cloud-based data processing latency mainly depends on the transmission delay of data to the cloud and the used data processing algorithm. To minimize the transmission delay, it is important to compress the transferred data without reducing the quality of the data. When using data compression algorithms, it is important to validate the impact of these algorithms on the detection quality. This work evaluates the effects of image compression and transmission over wireless interfaces on state of the art neural networks. Therefore, a modern image processing platform for next generation automotive processing architectures, as used in software defined vehicles, is introduced. The impacts of different image encoders as well as data transmission parameters are investigated and discussed.

关键词： image compression

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：