Pub Date : 2022-11-01DOI: 10.1109/CCISP55629.2022.9974302
Han Yang, Xueqin Jiang, H. Ge, Yuting Cao, Rong Ye
With the innovation and progress of data analysis, human action recognition has become a significant research direction with broad applications in many situations. We propose a skeleton temporal graph (STG) based on graph signal processing (GSP), graph spectral domain, and human action recognition. The temporal information between adjacent frames of action data is extracted by uniform sampling and redefining temporal edge weights. We reconstruct the graph Laplacian matrix from the skeleton and temporal information. According to the graph Laplacian matrix, the coefficient matrix used for classification is calculated by spectral graph wavelet transform (SGWT). In addition, we use a recent classification method eXtreme Gradient Boosting (XGBoost), to improve experimental accuracy. Our method outperforms the existing approach when applied to three publicly available datasets.
{"title":"Graph Construction Based on Temporal Information and Frame Sampling for Human Action Recognition","authors":"Han Yang, Xueqin Jiang, H. Ge, Yuting Cao, Rong Ye","doi":"10.1109/CCISP55629.2022.9974302","DOIUrl":"https://doi.org/10.1109/CCISP55629.2022.9974302","url":null,"abstract":"With the innovation and progress of data analysis, human action recognition has become a significant research direction with broad applications in many situations. We propose a skeleton temporal graph (STG) based on graph signal processing (GSP), graph spectral domain, and human action recognition. The temporal information between adjacent frames of action data is extracted by uniform sampling and redefining temporal edge weights. We reconstruct the graph Laplacian matrix from the skeleton and temporal information. According to the graph Laplacian matrix, the coefficient matrix used for classification is calculated by spectral graph wavelet transform (SGWT). In addition, we use a recent classification method eXtreme Gradient Boosting (XGBoost), to improve experimental accuracy. Our method outperforms the existing approach when applied to three publicly available datasets.","PeriodicalId":431851,"journal":{"name":"2022 7th International Conference on Communication, Image and Signal Processing (CCISP)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128722144","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 : 2022-11-01DOI: 10.1109/CCISP55629.2022.9974454
Jian-Xun Zhu, Jun-jing Tian, Jinqiang Ma, Zhi-Hong Miao, Zhanrong Guo, Lin Fan
Objective To visually display the research results of network security by knowledge mapping, analyze the research status of this field, and provide reference for future research in this field. Methods In this paper, 2197 articles in “CNKI” and 4932 articles in “WOS” database from 2012 to 2021 were visualized by literature metrology. CiteSpace software is used for metrological analysis and knowledge mapping. Results The research shows that the total number of network security documents in 2012–2021 shows an upward trend. There is less cooperation among the core authors. The US and China dominate the field of cyber security. There are both similarities and differences in network security research at home and abroad. Foreign research focuses on network attack and defense, while domestic research is widely distributed, including personal information, digital economy, artificial intelligence, political security and international governance. But the domestic research lacks the depth, and should strengthen the research on the cyber attack and defense.
{"title":"Visualization Analysis of Domestic and Foreign Cyber Security Research Based on CiteSpace","authors":"Jian-Xun Zhu, Jun-jing Tian, Jinqiang Ma, Zhi-Hong Miao, Zhanrong Guo, Lin Fan","doi":"10.1109/CCISP55629.2022.9974454","DOIUrl":"https://doi.org/10.1109/CCISP55629.2022.9974454","url":null,"abstract":"Objective To visually display the research results of network security by knowledge mapping, analyze the research status of this field, and provide reference for future research in this field. Methods In this paper, 2197 articles in “CNKI” and 4932 articles in “WOS” database from 2012 to 2021 were visualized by literature metrology. CiteSpace software is used for metrological analysis and knowledge mapping. Results The research shows that the total number of network security documents in 2012–2021 shows an upward trend. There is less cooperation among the core authors. The US and China dominate the field of cyber security. There are both similarities and differences in network security research at home and abroad. Foreign research focuses on network attack and defense, while domestic research is widely distributed, including personal information, digital economy, artificial intelligence, political security and international governance. But the domestic research lacks the depth, and should strengthen the research on the cyber attack and defense.","PeriodicalId":431851,"journal":{"name":"2022 7th International Conference on Communication, Image and Signal Processing (CCISP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128649403","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 : 2022-11-01DOI: 10.1109/CCISP55629.2022.9974475
Liu JinKua, Yang ChenXiang, H. B. Abdalla
Style transfer is a novel and successful technology in the field of computer vision which allow people to create art pieces without training. This research has combined style transfer, colorization, and super-resolution algorithm to create a method to create art pieces from the black-white image as content and sketch art pieces as style with high resolution. This method could significantly lower the demand for art creation and allow people with little artistic skill to create desired artwork. Moreover, this research compares the impact of the different parameters in style transfer and the influence of the colorization in different processing stages resulting that colorizing the content image before style transfer would create a single style image that is more controllable but colorizing the generated image after style transfer would result in a more unpredictable multi-style image which depends on the training dataset of the image and the ratio of the weight in style transfer.
{"title":"Enhanced style transfer with colorization and super-resolution","authors":"Liu JinKua, Yang ChenXiang, H. B. Abdalla","doi":"10.1109/CCISP55629.2022.9974475","DOIUrl":"https://doi.org/10.1109/CCISP55629.2022.9974475","url":null,"abstract":"Style transfer is a novel and successful technology in the field of computer vision which allow people to create art pieces without training. This research has combined style transfer, colorization, and super-resolution algorithm to create a method to create art pieces from the black-white image as content and sketch art pieces as style with high resolution. This method could significantly lower the demand for art creation and allow people with little artistic skill to create desired artwork. Moreover, this research compares the impact of the different parameters in style transfer and the influence of the colorization in different processing stages resulting that colorizing the content image before style transfer would create a single style image that is more controllable but colorizing the generated image after style transfer would result in a more unpredictable multi-style image which depends on the training dataset of the image and the ratio of the weight in style transfer.","PeriodicalId":431851,"journal":{"name":"2022 7th International Conference on Communication, Image and Signal Processing (CCISP)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132004219","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 : 2022-11-01DOI: 10.1109/CCISP55629.2022.9974419
Tiansheng Shen, Wenying Chen, Binbin Zhu, Yan Wang, Yongliang Zhou, Xiuping Wang
N owadays, indoor positioning algorithms have attracted comprehensive attention and research. Due to multipath effect in the complex indoor environment, it is difficult to position with high precision. In this paper a BLE(Bluetooth Low Energy) indoor positioning algorithm based on ISSA(Improved Sparrow Search Algorithm) with Cat chaotic mapping is proposed. Firstly the abnormal data are removed by BDOR(Bilateral Direction Outlier Removal) algorithm before establishing the database of the collected BLE data. Then the rest of data are filtered by improved Kalman filter. The ISSA-Elman models are studied to predict the horizontal and vertical coordinates of the point to be tested. Results of experiments reveal that the proposed algorithm performs precisely in indoor positioning. The minimal position error is nearly 2cm.
目前，室内定位算法得到了广泛的关注和研究。由于复杂的室内环境中存在多径效应，难以实现高精度定位。本文提出了一种基于Cat混沌映射的改进麻雀搜索算法(ISSA)的BLE(Bluetooth Low Energy)室内定位算法。首先用BDOR(Bilateral Direction Outlier Removal)算法去除异常数据，然后建立采集到的BLE数据数据库。然后用改进的卡尔曼滤波对剩余数据进行滤波。研究了ISSA-Elman模型来预测待测点的横坐标和纵坐标。实验结果表明，该算法在室内定位中具有较好的精度。最小位置误差接近2cm。
{"title":"An Indoor positioning method based on Improved Elman neural network using sparrow search","authors":"Tiansheng Shen, Wenying Chen, Binbin Zhu, Yan Wang, Yongliang Zhou, Xiuping Wang","doi":"10.1109/CCISP55629.2022.9974419","DOIUrl":"https://doi.org/10.1109/CCISP55629.2022.9974419","url":null,"abstract":"N owadays, indoor positioning algorithms have attracted comprehensive attention and research. Due to multipath effect in the complex indoor environment, it is difficult to position with high precision. In this paper a BLE(Bluetooth Low Energy) indoor positioning algorithm based on ISSA(Improved Sparrow Search Algorithm) with Cat chaotic mapping is proposed. Firstly the abnormal data are removed by BDOR(Bilateral Direction Outlier Removal) algorithm before establishing the database of the collected BLE data. Then the rest of data are filtered by improved Kalman filter. The ISSA-Elman models are studied to predict the horizontal and vertical coordinates of the point to be tested. Results of experiments reveal that the proposed algorithm performs precisely in indoor positioning. The minimal position error is nearly 2cm.","PeriodicalId":431851,"journal":{"name":"2022 7th International Conference on Communication, Image and Signal Processing (CCISP)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127854479","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 : 2022-11-01DOI: 10.1109/CCISP55629.2022.9974468
Yuanxu Liu, Jun Ma, X. Ren, V. Tsviatkou, Hao Li
We proposed a modified thinning framework that based on the scale space technique to automatically extract skeletons from images without manual-tuning. The proposed framework can increase the robustness of the thinning algorithm, it not only can suppress the boundary noise, but also can alleviate the inner noise. These two types of noise generally cause the appearance of the abundant of the unwanted branches in the outcome of the thinning algorithm, which arise the difficulties of the later recognition or matching process in skeleton. The experiment has proved the proposed framework has better performance when comparing with the other existing methods.
{"title":"A Modified Thinning Framework Against Noise","authors":"Yuanxu Liu, Jun Ma, X. Ren, V. Tsviatkou, Hao Li","doi":"10.1109/CCISP55629.2022.9974468","DOIUrl":"https://doi.org/10.1109/CCISP55629.2022.9974468","url":null,"abstract":"We proposed a modified thinning framework that based on the scale space technique to automatically extract skeletons from images without manual-tuning. The proposed framework can increase the robustness of the thinning algorithm, it not only can suppress the boundary noise, but also can alleviate the inner noise. These two types of noise generally cause the appearance of the abundant of the unwanted branches in the outcome of the thinning algorithm, which arise the difficulties of the later recognition or matching process in skeleton. The experiment has proved the proposed framework has better performance when comparing with the other existing methods.","PeriodicalId":431851,"journal":{"name":"2022 7th International Conference on Communication, Image and Signal Processing (CCISP)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126644020","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}
With the digital transformation of the power system, it is of great significance to realize intelligent identification of external breaking defects of overhead transmission lines and power pylons. This paper proposes a YOLOX-based detection method of triple-cascade feature level fusion for power system external defects. Based on YOLOX, the triple-cascade feature level fusion defect recognition and detection method, which is classified layer by layer according to the device inclusion relationship, are adopted. First, the types of equipment are judged, and then the grading standard are determined. Further, the part of the defect types, which are difficult to distinguish in the traditional machine learning algorithm, are refined and identified for the details. Finally, the proposed method is verified based on Python and NVIDIA Jetson TX2 platform with using the image data-set of the overhead transmission lines and power pylons. The mAP value of the model reaches 95.34%, which is higher 10.13% than that of YOLOX, and the detection speed reaches 32fps, which shows a promising performance for the robustness and real-time requirements of defect identification in the new power system.
{"title":"A YOLOX-Based Detection Method of Triple-Cascade Feature Level Fusion for Power System External Defects","authors":"Yufeng Sheng, Yingying Dai, Zixiao Luo, Chengming Jin, Chao Jiang, Liang Xue, Haoyang Cui","doi":"10.1109/CCISP55629.2022.9974428","DOIUrl":"https://doi.org/10.1109/CCISP55629.2022.9974428","url":null,"abstract":"With the digital transformation of the power system, it is of great significance to realize intelligent identification of external breaking defects of overhead transmission lines and power pylons. This paper proposes a YOLOX-based detection method of triple-cascade feature level fusion for power system external defects. Based on YOLOX, the triple-cascade feature level fusion defect recognition and detection method, which is classified layer by layer according to the device inclusion relationship, are adopted. First, the types of equipment are judged, and then the grading standard are determined. Further, the part of the defect types, which are difficult to distinguish in the traditional machine learning algorithm, are refined and identified for the details. Finally, the proposed method is verified based on Python and NVIDIA Jetson TX2 platform with using the image data-set of the overhead transmission lines and power pylons. The mAP value of the model reaches 95.34%, which is higher 10.13% than that of YOLOX, and the detection speed reaches 32fps, which shows a promising performance for the robustness and real-time requirements of defect identification in the new power system.","PeriodicalId":431851,"journal":{"name":"2022 7th International Conference on Communication, Image and Signal Processing (CCISP)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126788333","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 : 2022-11-01DOI: 10.1109/CCISP55629.2022.9974464
Peng Zou, Hongtu Xie, Xiao Hu, Jinfeng He, Guoqian Wang
The circular imaging technology of the building layout can obtain its internal information, which has been widely applied in the field of anti-terrorisms and combats. The problems of the imaging broadening caused by the scattering of the echo in the process of the penetrating between building walls bring the great inconvenience to the building layout reconstruction. In this paper, a multi-scale and multi-directional fusion method for the building circular layout image based on the morphological transformation has been proposed. Firstly, the multi-scale and multi-directional decomposition of the images from the various perspectives has been carried out by the nonsubsampled shearlet transformation. Moreover, the fusion method based on the morphological changes is carried out for low-frequency images, while the fusion method of taking the large regional variance is adopted for the high-frequency images. Finally, the inverse nonsubsampled shearlet transform transformation is respectively performed on the fused images, which can obtain the high-quality and high-precision building layout image. The experiment results are shown to prove the effectiveness of the proposed method.
{"title":"Multi-scale and Multi-directional Fusion Method for Building Circular Layout Image Based on Morphological Transformation","authors":"Peng Zou, Hongtu Xie, Xiao Hu, Jinfeng He, Guoqian Wang","doi":"10.1109/CCISP55629.2022.9974464","DOIUrl":"https://doi.org/10.1109/CCISP55629.2022.9974464","url":null,"abstract":"The circular imaging technology of the building layout can obtain its internal information, which has been widely applied in the field of anti-terrorisms and combats. The problems of the imaging broadening caused by the scattering of the echo in the process of the penetrating between building walls bring the great inconvenience to the building layout reconstruction. In this paper, a multi-scale and multi-directional fusion method for the building circular layout image based on the morphological transformation has been proposed. Firstly, the multi-scale and multi-directional decomposition of the images from the various perspectives has been carried out by the nonsubsampled shearlet transformation. Moreover, the fusion method based on the morphological changes is carried out for low-frequency images, while the fusion method of taking the large regional variance is adopted for the high-frequency images. Finally, the inverse nonsubsampled shearlet transform transformation is respectively performed on the fused images, which can obtain the high-quality and high-precision building layout image. The experiment results are shown to prove the effectiveness of the proposed method.","PeriodicalId":431851,"journal":{"name":"2022 7th International Conference on Communication, Image and Signal Processing (CCISP)","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122636826","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 : 2022-11-01DOI: 10.1109/CCISP55629.2022.9974528
Yiru Niu, Hong Jiang, Xiaobing Huang, Qi Zhao, H. Xiang, Xiaofeng Xia, Yue Zhao
With the introduction of the smart city concept and the rapid development of various emerging technologies such as cloud computing, Internet of Things, big data and artificial intelligence, traditional parks have also ushered in new opportunities and are developing in a smarter direction. Since the park is an open area with frequent movement of people and equipment, it is prone to security problems. At this point, logs will be a good aid for incident accountability. In this paper, we mainly use Merkle hash tree and distributed storage for secure storage of logs. Meanwhile, to ensure the non-repudiation of logs, the signatures of users and log-generating devices are required when generating logs. After analysis, the scheme proposed in this paper can achieve tamper-proof and traceability of logs.
{"title":"Tamper-Proof Storage of User Movement Logs in Smart Parks","authors":"Yiru Niu, Hong Jiang, Xiaobing Huang, Qi Zhao, H. Xiang, Xiaofeng Xia, Yue Zhao","doi":"10.1109/CCISP55629.2022.9974528","DOIUrl":"https://doi.org/10.1109/CCISP55629.2022.9974528","url":null,"abstract":"With the introduction of the smart city concept and the rapid development of various emerging technologies such as cloud computing, Internet of Things, big data and artificial intelligence, traditional parks have also ushered in new opportunities and are developing in a smarter direction. Since the park is an open area with frequent movement of people and equipment, it is prone to security problems. At this point, logs will be a good aid for incident accountability. In this paper, we mainly use Merkle hash tree and distributed storage for secure storage of logs. Meanwhile, to ensure the non-repudiation of logs, the signatures of users and log-generating devices are required when generating logs. After analysis, the scheme proposed in this paper can achieve tamper-proof and traceability of logs.","PeriodicalId":431851,"journal":{"name":"2022 7th International Conference on Communication, Image and Signal Processing (CCISP)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131291354","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 : 2022-11-01DOI: 10.1109/CCISP55629.2022.9974442
Dong Yun, Yang Yong, Xu Kaijun, Xuan ZhaoYu
ADS-B technology is one of the critical surveillance technologies identified by ICAO for the future and one of the four main directions that the China Civil Aviation New Technology Application and Development Committee has focused on shortly. According to the direction of aircraft data transmission, the airborne ADS-B application functions can be divided into OUT and IN categories. ADS-B IN means that the aircraft can receive ADS-B OUT information sent by other aircraft to provide operational support or information sent by ground station equipment to provide operational support to the pilot. ADS-B IN technology enables aircraft to receive ADS-B information from surrounding aircraft, achieve traffic situation awareness, and implement “TCAS-like” functions applicable to general aviation flight characteristics to improve general aviation flight safety further. This paper provides an in-depth analysis of the current status of international ADS-B IN technology development, discusses in detail the update of relevant policies and regulations of Chinese civil aviation and the progress of demonstration and validation, and analyzes the main problems existing in actual operation.
{"title":"Progress of ADS-B IN technology in civil aviation applications","authors":"Dong Yun, Yang Yong, Xu Kaijun, Xuan ZhaoYu","doi":"10.1109/CCISP55629.2022.9974442","DOIUrl":"https://doi.org/10.1109/CCISP55629.2022.9974442","url":null,"abstract":"ADS-B technology is one of the critical surveillance technologies identified by ICAO for the future and one of the four main directions that the China Civil Aviation New Technology Application and Development Committee has focused on shortly. According to the direction of aircraft data transmission, the airborne ADS-B application functions can be divided into OUT and IN categories. ADS-B IN means that the aircraft can receive ADS-B OUT information sent by other aircraft to provide operational support or information sent by ground station equipment to provide operational support to the pilot. ADS-B IN technology enables aircraft to receive ADS-B information from surrounding aircraft, achieve traffic situation awareness, and implement “TCAS-like” functions applicable to general aviation flight characteristics to improve general aviation flight safety further. This paper provides an in-depth analysis of the current status of international ADS-B IN technology development, discusses in detail the update of relevant policies and regulations of Chinese civil aviation and the progress of demonstration and validation, and analyzes the main problems existing in actual operation.","PeriodicalId":431851,"journal":{"name":"2022 7th International Conference on Communication, Image and Signal Processing (CCISP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131489056","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 : 2022-11-01DOI: 10.1109/CCISP55629.2022.9974317
Yingjiang Li, Yaping Deng, Yuzhong Zhong
This paper presents a multi-step stereo matching algorithm that can be applied to multiple scenes. To adapt to different application scenarios, the algorithm divides the stereo matching process into three steps: point, fragment, and plane. First, the texture points of an image are extracted and the stereo matching (point disparity) of these points is performed using the improved self-aware matching measure (SAMM) algorithm. Then, according to the edge information of the image, a smooth region is divided into different fragments in the horizontal direction. The disparity estimation of the smooth region (segment disparity) is obtained through the confidence propagation of disparity values of texture points in the fragments. Finally, based on the similarity of plane disparity, a disparity map is generated using the disparity refining algorithm (plane disparity), and a final high-precision disparity value is obtained. The experimental results show that the proposed algorithm has high operational efficiency and accurate disparity estimation. Moreover, the algorithm may be adapted for more application scenarios.
{"title":"Multi-step stereo matching algorithm based on image fragment and confidence propagation","authors":"Yingjiang Li, Yaping Deng, Yuzhong Zhong","doi":"10.1109/CCISP55629.2022.9974317","DOIUrl":"https://doi.org/10.1109/CCISP55629.2022.9974317","url":null,"abstract":"This paper presents a multi-step stereo matching algorithm that can be applied to multiple scenes. To adapt to different application scenarios, the algorithm divides the stereo matching process into three steps: point, fragment, and plane. First, the texture points of an image are extracted and the stereo matching (point disparity) of these points is performed using the improved self-aware matching measure (SAMM) algorithm. Then, according to the edge information of the image, a smooth region is divided into different fragments in the horizontal direction. The disparity estimation of the smooth region (segment disparity) is obtained through the confidence propagation of disparity values of texture points in the fragments. Finally, based on the similarity of plane disparity, a disparity map is generated using the disparity refining algorithm (plane disparity), and a final high-precision disparity value is obtained. The experimental results show that the proposed algorithm has high operational efficiency and accurate disparity estimation. Moreover, the algorithm may be adapted for more application scenarios.","PeriodicalId":431851,"journal":{"name":"2022 7th International Conference on Communication, Image and Signal Processing (CCISP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131741510","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：