检索结果-内蒙古大学图书馆

International Conference on Fuzzy Systems and Knowledge Discovery (FSKD)

作者： Guishan Zhang Haitao Deng Yaowen Chen Zhiwei Shen Renhua Wu Key Lab. of Digital Signal and Image Processing of Guangdong Province Shantou University Shantou China Department of Medical Imaging 2nd Affiliated Hospital Medical College of Shantou University Shantou China

ISBN: (纸本)9781467376839

Compressed sensing (CS) has been applied to the field of sub-sampled magnetic resonance imaging (MRI) reconstruction (CS-MRI). Fast iterative shrinkage-thresholding algorithm (FISTA) is an effective method for CS-MR images reconstruction. To investigate the accuracy and efficiency of the proposed algorithm, we applied it to the under-sampling MR images gained by different MRI scanning sequences. We found the peak signal to noise ratio (PSNR) of reconstructed MRI with varying sampling ratios diminished from Axial T1 weighted images (Ax T1) (45.13±12.97 dB), Axial T2 weighted images (Ax T2) (42.8 ± 14.95 dB), FLAIR (41.74 ± 14.15 dB), Diffusion Weighted Imaging (DWI) (40.23 ±17.40 dB) and Sagittal T2 weighted images (Sag T2) (36.28±12.32 dB), but there was no significant difference among the groups. In addition, the changes to the elapsed time of them was minor, Ax T1 (1.09±0.13 s), Ax T2 (1.30±0.13s), FLAIR (1.02±0.12s), DWI (1.07±0.13s) and Sag T2 (1.12±0.07s). Our results confirmed the stability of the proposed fast MRI reconstruction method for different scanning sequences. Further efforts are still needed to design the clinical sequence with sub-sampled acquisition strategy which may be a developing technique with clinical value.

关键词： Compressed Sensing Magnetic Resonance Imaging PSNR Fast Iterative Shrinkage-thresholding Algorithm (FISTA)

来源：评论

学校读者我要写书评

暂无评论

Understanding the Robustness of 3D Object Detection with Bird'View Representations in Autonomous Driving

Understanding the Robustness of 3D Object Detection with Bir...

引用

2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023

作者： Zhu, Zijian Zhang, Yichi Chen, Hai Dong, Yinpeng Zhao, Shu Ding, Wenbo Zhong, Jiachen Zheng, Shibao Institute of Image Communication and Network Engineering Shanghai Jiao Tong University China Institute for Ai Tsinghua University BNRist Center Thbi Lab Dept. of Comp. Sci. and Tech. China School of Computer Science and Technology Anhui University Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Information Materials and Intelligent Sensing Laboratory of Anhui Province China Saic Motor Ai Lab Zhongguancun Laboratory China

ISBN: (纸本)9798350301298

3D object detection is an essential perception task in autonomous driving to understand the environments. The Bird's-Eye-View (BEV) representations have significantly improved the performance of 3D detectors with camera inputs on popular benchmarks. However, there still lacks a systematic understanding of the robustness of these vision-dependent BEV models, which is closely related to the safety of autonomous driving systems. In this paper, we evaluate the natural and adversarial robustness of various representative models under extensive settings, to fully understand their behaviors influenced by explicit BEV features compared with those without BEV. In addition to the classic settings, we propose a 3D consistent patch attack by applying adversarial patches in the 3D space to guarantee the spatiotemporal consistency, which is more realistic for the scenario of autonomous driving. With substantial experiments, we draw several findings: 1) BEV models tend to be more stable than previous methods under different natural conditions and common corruptions due to the expressive spatial representations;2) BEV models are more vulnerable to adversarial noises, mainly caused by the redundant BEV features;3) Camera-LiDARfusion models have superior performance under different settings with multi-modal inputs, but BEV fusion model is still vulnerable to adversarial noises of both point cloud and image. These findings alert the safety issue in the applications of BEV detectors and could facilitate the development of more robust models. © 2023 IEEE.

关键词： Autonomous driving

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Feature-based Quantization Algorithm for Point Cloud Compression

An Adaptive Feature-based Quantization Algorithm for Point C...

引用

Picture Coding Symposium, PCS

作者： Da Ai Hongying Lu Yurong Yang Ying Liu National Key Lab. of Electronic Information Processing for Applications in Crime Scene Investigation Ministry of Public Security Xi‘an China Center for Image and Information Processing Xi'an University of Posts and Telecommunications Xi'an China The International Joint-Research Center for Wireless Communication and Information Processing Xi‘an University of Posts and Telecommunications Xi‘an China

To reduce over-rasterization distortion caused by global uniform quantization for static surface point cloud, an adaptive quantization coding method based on feature mining is proposed. Combining spatial position and texture feature of point clouds with level of details, the quantization increment is dynamically set according to feature priority, which can reserve the number of effective points to the maximum extent, and reduce the rasterization distortion. Experimental results show that the proposed method can effectively enhance the subjective reconstruction quality of compressed point cloud, gaining better results of rate-distortion optimization.

关键词： Graphics Geometry Three-dimensional displays Quantization (signal) Transform coding Rate-distortion Distortion

来源：评论

学校读者我要写书评

暂无评论

image Fusion Algorithm Based on Features Motivated Multi-Channel Pulse Coupled Neural Networks

Image Fusion Algorithm Based on Features Motivated Multi-Cha...

引用

International Conference on Bioinformatics and Biomedical Engineering (ICBBE)

作者： Xiaobo Qu Jingwen Yan 1Dept. of Communication Engineering Xiamen University.Fujian P.R.China 361005 1Dept. of Communication Engineering Xiamen University.Fujian P.R.China 3610052Key Lab of Digital Signal and Image Processing ofGuangdong Province Shantou University.Guangdong P.R.China 515063.

Pulse coupled neural networks (PCNN) is a mammal visual cortex-inspired artificial neural networks. Owing to the coupling links in neurons, PCNN is successful to utilize the local information, thus it has been successfully employed in image fusion. However, in traditional PCNN for image fusion, value of per pixel is used to motivate per neuron. In this paper, image feature of per pixel, e.g. gradient and local energy, is used to motivate per neuron and generate firing maps. Each firing map is corresponding to one type feature. Furthermore, a new multi-channel PCNN is presented to combine these firing maps via a weighting function which measures the contribution of these features to the fused image quality. Finally, pixels with maximum firing times, when firing times of source images are compared, are selected as the pixels of the fused image. Experimental results demonstrate that the proposed algorithm outperforms Wavelet- based and Wavelet-PCNN-based fusion algorithms.

关键词： image fusion Neural networks Neurons Pixel Discrete wavelet transforms image processing Artificial neural networks Fusion power generation Humans Pulse modulation

来源：评论

学校读者我要写书评

暂无评论

Fast multiple reference frame selection algorithm based on H.264/AVC

引用

Sichuan Daxue Xuebao (Gongcheng Kexue Ban)/Journal of Sichuan University (Engineering Science Edition) 2005年第4期37卷 100-103页

作者： Teng, Guo-Wei Zhang, Zhao-Yang Zhang, Yi-Jun Zhang, Wen-Jun Key Lab. of Advanced Display and System Appl. Shanghai Univ. Shanghai 200072 China Inst. of Image Communication and Info. Processing Shanghai Jiaotong Univ. Shanghai 200030 China

A reference frame selection algorithm was proposed to fast determine the reference frames required by the current macroblock, which can avoid redundant reference frames. This algorithm can utilize the texture characteristic of sequence to estimate the possible number of reference frames, and further determine the best reference frame of 16 × 16, 8 × 8 and 4 × 4 modes according to the motion information and the monotony of RDO among different reference frames. So the likely reference frames of other modes may be adaptively chosen according to the correlation of reference frame selection with the above modes. The results showed that it can save about 30% of encoding time while maintaining nearly unchanged PSNR and bit-rate.

关键词： Video signal processing

来源：评论

学校读者我要写书评

暂无评论

Understanding the Robustness of 3D Object Detection with Bird'View Representations in Autonomous Driving

Understanding the Robustness of 3D Object Detection with Bir...

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Zijian Zhu Yichi Zhang Hai Chen Yinpeng Dong Shu Zhao Wenbo Ding Jiachen Zhong Shibao Zheng Institute of Image Communication and Network Engineering Shanghai Jiao Tong University Dept. of Comp. Sci. and Tech. THBI Lab Institute for AI Tsinghua University BNRist Center Zhongguancun Laboratory Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Information Materials and Intelligent Sensing Laboratory of Anhui Province School of Computer Science and Technology Anhui University SAIC Motor AI Lab

3D object detection is an essential perception task in autonomous driving to understand the environments. The Bird's-Eye-View (BEV) representations have significantly improved the performance of 3D detectors with camera inputs on popular benchmarks. However, there still lacks a systematic understanding of the robustness of these vision-dependent BEV models, which is closely related to the safety of autonomous driving systems. In this paper, we evaluate the natural and adversarial robustness of various representative models under extensive settings, to fully understand their behaviors influenced by explicit BEV features compared with those without BEV. In addition to the classic settings, we propose a 3D consistent patch attack by applying adversarial patches in the 3D space to guarantee the spatiotemporal consistency, which is more realistic for the scenario of autonomous driving. With substantial experiments, we draw several findings: 1) BEV models tend to be more stable than previous methods under different natural conditions and common corruptions due to the expressive spatial representations; 2) BEV models are more vulnerable to adversarial noises, mainly caused by the redundant BEV features; 3) Camera-LiDARfusion models have superior performance under different settings with multi-modal inputs, but BEV fusion model is still vulnerable to adversarial noises of both point cloud and image. These findings alert the safety issue in the applications of BEV detectors and could facilitate the development of more robust models.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Understanding the Robustness of 3D Object Detection with Bird’s-Eye-View Representations in Autonomous Driving

arXiv

引用

arXiv 2023年

作者： Zhu, Zijian Zhang, Yichi Chen, Hai Dong, Yinpeng Zhao, Shu Ding, Wenbo Zhong, Jiachen Zheng, Shibao Institute of Image Communication and Network Engineering Shanghai Jiao Tong University China Dept. of Comp. Sci. and Tech. Institute for AI THBI Lab BNRist Center Tsinghua University China Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education School of Computer Science and Technology Anhui University Information Materials and Intelligent Sensing Laboratory of Anhui Province China SAIC Motor AI Lab China Zhongguancun Laboratory China

3D object detection is an essential perception task in autonomous driving to understand the environments. The Bird’s-Eye-View (BEV) representations have significantly improved the performance of 3D detectors with camera inputs on popular benchmarks. However, there still lacks a systematic understanding of the robustness of these vision-dependent BEV models, which is closely related to the safety of autonomous driving systems. In this paper, we evaluate the natural and adversarial robustness of various representative models under extensive settings, to fully understand their behaviors influenced by explicit BEV features compared with those without BEV. In addition to the classic settings, we propose a 3D consistent patch attack by applying adversarial patches in the 3D space to guarantee the spatiotemporal consistency, which is more realistic for the scenario of autonomous driving. With substantial experiments, we draw several findings: 1) BEV models tend to be more stable than previous methods under different natural conditions and common corruptions due to the expressive spatial representations;2) BEV models are more vulnerable to adversarial noises, mainly caused by the redundant BEV features;3) Camera-LiDAR fusion models have superior performance under different settings with multi-modal inputs, but BEV fusion model is still vulnerable to adversarial noises of both point cloud and image. These findings alert the safety issue in the applications of BEV detectors and could facilitate the development of more robust models. Code availab.e at https: //***/zzj403/BEV_Robust. Copyright © 2023, The Authors. All rights reserved.

关键词： Birds

来源：评论

学校读者我要写书评

暂无评论

Millimeter-wave radar image analysis for the traffic sensing

引用

Journal of Physics: Conference Series 2020年第5期1507卷

作者： H M Wu F Qi J K Wang Y Wang School of communication science and engineering Northeastern University Liaoning Province Shenyang 110819 China Shenyang Institute of Automation Chinese Academy of Sciences China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences China Key Laboratory of Opto-Electronic Information Processing Chinese Academy of Sciences China Key Lab of Image Understanding and Computer Vision Liaoning Province Shenyang 110016 China

Millimeter-wave(MMW) radar sensing is one of the most promising technologies to provide safe navigation for autonomous vehicles due to its expected high-resolution imaging capability However, driverless cars have higher request for different environment and light conditions. Therefore, millimetre-wave imaging is of paramount importance for complex load scenario. In this paper, we have built models of pavement pits and bulges and analysed their with differences ways of antennas. A comparison of the imaging performance of experimental systems operating at a MMW radar and a Lidar is presented with the analysis of features for initial image interpretation Experimental images of the complex road surface are made by a 94GHz frequency-modulated continuous-wave (FMCW) radar technique with 3mm wavelength.

关键词：

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：