Pub Date : 2021-11-01DOI: 10.3724/sp.j.1089.2021.18788
Yunlong Che, Yue Qi
{"title":"A Survey on Depth Based Hand Pose Estimation","authors":"Yunlong Che, Yue Qi","doi":"10.3724/sp.j.1089.2021.18788","DOIUrl":"https://doi.org/10.3724/sp.j.1089.2021.18788","url":null,"abstract":"","PeriodicalId":52442,"journal":{"name":"计算机辅助设计与图形学学报","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48734588","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
To solve the problems of mistaken coloring and color bleeding in the current colorization methods, an end-to-end deep neural network is proposed to achieve remote sensing image colorization. First, the multi-scale residual receptive filed net is introduced to extract the key features of source image. Second, a color information recovery network is con-structed by using U-Net, complex residual structure, attention mechanism, sequeeze-and-excitation and pixel-shuffle blocks to obtain color result. NWPU-RESISC45 dataset is chosen for model training and validation. Compared with other color methods, the PSNR value of the proposed method is increased by 6-10 dB on average and the SSIM value is increased by 0.05-0.11. In addition, the proposed method also achieves satisfactory color results on RSSCN7 and AID datasets.
{"title":"Remote Sensing Image Colorization Based on Deep Neural Networks with Multi-Scale Residual Receptive Filed","authors":"Jianan Feng, Qian Jiang, Xin Jin, Shin-Jye Lee, Shanshan Huang, Shao-qing Yao","doi":"10.3724/sp.j.1089.2021.18747","DOIUrl":"https://doi.org/10.3724/sp.j.1089.2021.18747","url":null,"abstract":"To solve the problems of mistaken coloring and color bleeding in the current colorization methods, an end-to-end deep neural network is proposed to achieve remote sensing image colorization. First, the multi-scale residual receptive filed net is introduced to extract the key features of source image. Second, a color information recovery network is con-structed by using U-Net, complex residual structure, attention mechanism, sequeeze-and-excitation and pixel-shuffle blocks to obtain color result. NWPU-RESISC45 dataset is chosen for model training and validation. Compared with other color methods, the PSNR value of the proposed method is increased by 6-10 dB on average and the SSIM value is increased by 0.05-0.11. In addition, the proposed method also achieves satisfactory color results on RSSCN7 and AID datasets.","PeriodicalId":52442,"journal":{"name":"计算机辅助设计与图形学学报","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41875528","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-01DOI: 10.3724/sp.j.1089.2021.18799
Yuyang Xie, Lingjie Li, Wenjian Yu
: Circuit simulation becomes more and more important in integrated circuit design. For VLSI circuits, the simulation usually outputs signal waveforms occupying massive storage space. The compression of these signal waveforms becomes crucial to the efficiency of circuit simulation. Logic simulation mainly outputs the signal values at the time of signal transition and some auxiliary information such as signal name, signal type, signal width. A compression method for auxiliary information is proposed. Then, the signal name compression scheme in existing work is improved according to the characteristics of signal value data, and a more efficient digital waveform compression storage format is proposed. The proposed format is more adaptive to the variable-length coding for compression. At the same time, general compression algorithms can be used for secondary compression, thereby further
{"title":"Efficient Digital Waveform Compression Method for Logic Simulation of Integrated Circuits","authors":"Yuyang Xie, Lingjie Li, Wenjian Yu","doi":"10.3724/sp.j.1089.2021.18799","DOIUrl":"https://doi.org/10.3724/sp.j.1089.2021.18799","url":null,"abstract":": Circuit simulation becomes more and more important in integrated circuit design. For VLSI circuits, the simulation usually outputs signal waveforms occupying massive storage space. The compression of these signal waveforms becomes crucial to the efficiency of circuit simulation. Logic simulation mainly outputs the signal values at the time of signal transition and some auxiliary information such as signal name, signal type, signal width. A compression method for auxiliary information is proposed. Then, the signal name compression scheme in existing work is improved according to the characteristics of signal value data, and a more efficient digital waveform compression storage format is proposed. The proposed format is more adaptive to the variable-length coding for compression. At the same time, general compression algorithms can be used for secondary compression, thereby further","PeriodicalId":52442,"journal":{"name":"计算机辅助设计与图形学学报","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44610227","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-01DOI: 10.3724/sp.j.1089.2021.18780
Kexin Meng, Meng-Zhu Liu, S. Guo, Shuli Mei
{"title":"Hermite-Shannon-Cosine Interval Wavelet and Its Application in Adaptive Distribute Interpolation on Curves","authors":"Kexin Meng, Meng-Zhu Liu, S. Guo, Shuli Mei","doi":"10.3724/sp.j.1089.2021.18780","DOIUrl":"https://doi.org/10.3724/sp.j.1089.2021.18780","url":null,"abstract":"","PeriodicalId":52442,"journal":{"name":"计算机辅助设计与图形学学报","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46655161","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-01DOI: 10.3724/sp.j.1089.2021.18789
Xueyi Ye, Tao Wang, Na Ying, Dingwei Qian
Pointing to weak robustness caused by the noise sensitivity and feature redundancy of present face recognition methods with high-order features, a new method of the local high-order principal direction pattern based on “gradient face” is proposed. Firstly, the gradient face convolution operator designed is used to compute the sum of multi-directional gradient components of pixels to construct a gradient face. Then, the principal direction grouping strategy is introduced on the gradient face to characterize its high-order derivative features, and a principal direction feature map is formed according to the feature code of high-order derivatives direction changes in local neighborhood. Finally, block statistics and cascading of histogram features are made a vector to be input in to a support vector machine for multi-classification. Experimental results of several public face databases show that the proposed method is robust to changes in illumination, expression, and facial occlusion and has higher recognition efficiency.
{"title":"Face Recognition with Local High-Order Principal Direction Pattern Based on “Gradient Face”","authors":"Xueyi Ye, Tao Wang, Na Ying, Dingwei Qian","doi":"10.3724/sp.j.1089.2021.18789","DOIUrl":"https://doi.org/10.3724/sp.j.1089.2021.18789","url":null,"abstract":"Pointing to weak robustness caused by the noise sensitivity and feature redundancy of present face recognition methods with high-order features, a new method of the local high-order principal direction pattern based on “gradient face” is proposed. Firstly, the gradient face convolution operator designed is used to compute the sum of multi-directional gradient components of pixels to construct a gradient face. Then, the principal direction grouping strategy is introduced on the gradient face to characterize its high-order derivative features, and a principal direction feature map is formed according to the feature code of high-order derivatives direction changes in local neighborhood. Finally, block statistics and cascading of histogram features are made a vector to be input in to a support vector machine for multi-classification. Experimental results of several public face databases show that the proposed method is robust to changes in illumination, expression, and facial occlusion and has higher recognition efficiency.","PeriodicalId":52442,"journal":{"name":"计算机辅助设计与图形学学报","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41568329","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-01DOI: 10.3724/sp.j.1089.2021.18744
Shen Hou, Jinglong Li, Hailong Liu, Shaoqing Li, Yang Guo
To solve the problem that the physical unclonable function (PUF) structure is simple and vulnerable to modeling attacks, a strong PUF anti-attack obfuscation design based on linear feedback shift register (LFSR) is proposed. First, a fixed structure LFSR is used as a pseudo-random number generator to provide a random selection signal for the obfuscation logic. Then, a dynamic LFSR with multiple feedback polynomials is used as the obfuscation logic to obfuscate origin challenges. Finally, obfuscated challenges are loaded into the embedded PUF circuit so that the attacker cannot obtain real challenges. It improves the resistance of the PUF to modeling attacks. The proposed design is simulated by Python and FPGA. Experiments on the collected dataset show that the proposed PUF has ideal uniformity (49.8%) and uniqueness (49.9%) and keeps the same reliability. It has simple architecture and low hardware overhead and can resist a variety of modeling attacks including machine learning and deep learning.
{"title":"Design of Lightweight and Configurable Strong Physical Unclonable Function","authors":"Shen Hou, Jinglong Li, Hailong Liu, Shaoqing Li, Yang Guo","doi":"10.3724/sp.j.1089.2021.18744","DOIUrl":"https://doi.org/10.3724/sp.j.1089.2021.18744","url":null,"abstract":"To solve the problem that the physical unclonable function (PUF) structure is simple and vulnerable to modeling attacks, a strong PUF anti-attack obfuscation design based on linear feedback shift register (LFSR) is proposed. First, a fixed structure LFSR is used as a pseudo-random number generator to provide a random selection signal for the obfuscation logic. Then, a dynamic LFSR with multiple feedback polynomials is used as the obfuscation logic to obfuscate origin challenges. Finally, obfuscated challenges are loaded into the embedded PUF circuit so that the attacker cannot obtain real challenges. It improves the resistance of the PUF to modeling attacks. The proposed design is simulated by Python and FPGA. Experiments on the collected dataset show that the proposed PUF has ideal uniformity (49.8%) and uniqueness (49.9%) and keeps the same reliability. It has simple architecture and low hardware overhead and can resist a variety of modeling attacks including machine learning and deep learning.","PeriodicalId":52442,"journal":{"name":"计算机辅助设计与图形学学报","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47243132","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-01DOI: 10.3724/sp.j.1089.2021.18416
Sheng-sheng Wang, S. Lu, Bin Cao
{"title":"Medical Image Object Detection Algorithm for Privacy-Preserving Federated Learning","authors":"Sheng-sheng Wang, S. Lu, Bin Cao","doi":"10.3724/sp.j.1089.2021.18416","DOIUrl":"https://doi.org/10.3724/sp.j.1089.2021.18416","url":null,"abstract":"","PeriodicalId":52442,"journal":{"name":"计算机辅助设计与图形学学报","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47189192","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：