Pub Date : 2020-12-22DOI: 10.12086/OEE.2020.200111
Long Xiao, Bao Hua, Rao Changhui, Gao Guoqing, Zhou Luchun
Aiming at the shortcoming of low serial operational efficiency in the quality-map-guided phase-unwrapping algorithm proposed by Miguel, an improved algorithm for parallel merging of multiple low-reliability blocks is pro-posed. Under the condition that the original algorithm design idea is satisfied, the unwrapping path is redefined as the largest reliable edge of the block. In addition, based on the non-continuous characteristic of the unwrapping path of the original algorithm, a low-reliability block out-of-order merging strategy is proposed to make multiple merging tasks can be performed simultaneously. The improved algorithm uses a multi-threaded software architecture. The main thread is responsible for looping through the unprocessed blocks to check whether they meet the requirements of merging, and the child threads receive and perform the merge tasks. The experimental results show that the improved method is completely consistent with the processing results of the original algorithm, and the parallel improvement strategy can effectively use the computer's multi-core resources, so that the operational efficiency of the phase unwrapping algorithm is improved by more than 50%.
{"title":"Improved fast phase unwrapping algorithm based on parallel acceleration","authors":"Long Xiao, Bao Hua, Rao Changhui, Gao Guoqing, Zhou Luchun","doi":"10.12086/OEE.2020.200111","DOIUrl":"https://doi.org/10.12086/OEE.2020.200111","url":null,"abstract":"Aiming at the shortcoming of low serial operational efficiency in the quality-map-guided phase-unwrapping algorithm proposed by Miguel, an improved algorithm for parallel merging of multiple low-reliability blocks is pro-posed. Under the condition that the original algorithm design idea is satisfied, the unwrapping path is redefined as the largest reliable edge of the block. In addition, based on the non-continuous characteristic of the unwrapping path of the original algorithm, a low-reliability block out-of-order merging strategy is proposed to make multiple merging tasks can be performed simultaneously. The improved algorithm uses a multi-threaded software architecture. The main thread is responsible for looping through the unprocessed blocks to check whether they meet the requirements of merging, and the child threads receive and perform the merge tasks. The experimental results show that the improved method is completely consistent with the processing results of the original algorithm, and the parallel improvement strategy can effectively use the computer's multi-core resources, so that the operational efficiency of the phase unwrapping algorithm is improved by more than 50%.","PeriodicalId":39552,"journal":{"name":"Guangdian Gongcheng/Opto-Electronic Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90661397","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 : 2020-12-22DOI: 10.12086/OEE.2020.190636
Z. Baohua, Zhu Siyu, Lv Xiaoqi, Gu Yu, Wang Yueming, Liu Xin, Ren Yan, Li Jianjun, Zhang Ming
When the microLED is in the forward working direction, it is difficult to precisely adjust its voltage to obtain different brightness. Moreover, when the microLED/OLED is turned on, they will be in a closed state for a long time, causing the image display brightness to be deteriorated by the human eye. In order to solve these problems, this paper proposes a dual-frame decentralized fusion scanning strategy to achieve different brightness by adjusting the microLED/OLED on-time. Firstly, the method de-weights the data bits and inserts their on-times into the closed time. Then the data bit weights are double-frame fused after decentralization. Finally, the scanning order of the data bits is redefined. According to the proposed scanning strategy, we designed a scanning controller to drive digital on-silicon microdisplay. The results show that the dual-frame decentralized fusion scan proposed in this paper can accurately adjust the luminance of microLED/OLED and improve the brightness of the image observed by human eyes. Compared with other scanning strategies, the scanning strategy improves the scanning efficiency to 93.75%, the field frequency is increased to 2040 Hz, the scanning clock frequency is 102.36 MHz, and the scanning data bandwidth is reduced. The feasibility of the scan controller is proved by testing at last.
{"title":"Soft multilabel learning and deep feature fusion for unsupervised person re-identification","authors":"Z. Baohua, Zhu Siyu, Lv Xiaoqi, Gu Yu, Wang Yueming, Liu Xin, Ren Yan, Li Jianjun, Zhang Ming","doi":"10.12086/OEE.2020.190636","DOIUrl":"https://doi.org/10.12086/OEE.2020.190636","url":null,"abstract":"When the microLED is in the forward working direction, it is difficult to precisely adjust its voltage to obtain different brightness. Moreover, when the microLED/OLED is turned on, they will be in a closed state for a long time, causing the image display brightness to be deteriorated by the human eye. In order to solve these problems, this paper proposes a dual-frame decentralized fusion scanning strategy to achieve different brightness by adjusting the microLED/OLED on-time. Firstly, the method de-weights the data bits and inserts their on-times into the closed time. Then the data bit weights are double-frame fused after decentralization. Finally, the scanning order of the data bits is redefined. According to the proposed scanning strategy, we designed a scanning controller to drive digital on-silicon microdisplay. The results show that the dual-frame decentralized fusion scan proposed in this paper can accurately adjust the luminance of microLED/OLED and improve the brightness of the image observed by human eyes. Compared with other scanning strategies, the scanning strategy improves the scanning efficiency to 93.75%, the field frequency is increased to 2040 Hz, the scanning clock frequency is 102.36 MHz, and the scanning data bandwidth is reduced. The feasibility of the scan controller is proved by testing at last.","PeriodicalId":39552,"journal":{"name":"Guangdian Gongcheng/Opto-Electronic Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88338174","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 : 2020-12-22DOI: 10.12086/OEE.2020.190688
Huang Siyuan, L. Limin, Dong Jian, Fu Xiongjun
LiDAR plays an important role in the field of unmanned driving. Ground filtering is the key technology to separate and extract the ground information from the point cloud data acquired by LiDAR. Firstly, the development and classification of vehicle LiDAR scans (VLS) are introduced, and the advantages and disadvantages of all kinds of VLS are discussed. Then, the development of VLS ground filtering algorithm is studied and classified. The evaluation methods and standards of ground filtering accuracy are described, and three typical algorithms are compared and analyzed. Finally, the shortcomings of current VLS and its ground filtering algorithms are summarized, and the future development trend is prospected.
{"title":"Review of ground filtering algorithms for vehicle LiDAR scans point cloud dataReview of ground filtering algorithms for vehicle LiDAR scans point cloud data","authors":"Huang Siyuan, L. Limin, Dong Jian, Fu Xiongjun","doi":"10.12086/OEE.2020.190688","DOIUrl":"https://doi.org/10.12086/OEE.2020.190688","url":null,"abstract":"LiDAR plays an important role in the field of unmanned driving. Ground filtering is the key technology to separate and extract the ground information from the point cloud data acquired by LiDAR. Firstly, the development and classification of vehicle LiDAR scans (VLS) are introduced, and the advantages and disadvantages of all kinds of VLS are discussed. Then, the development of VLS ground filtering algorithm is studied and classified. The evaluation methods and standards of ground filtering accuracy are described, and three typical algorithms are compared and analyzed. Finally, the shortcomings of current VLS and its ground filtering algorithms are summarized, and the future development trend is prospected.","PeriodicalId":39552,"journal":{"name":"Guangdian Gongcheng/Opto-Electronic Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86700013","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 : 2020-12-22DOI: 10.12086/OEE.2020.200067
Zhang Xinli, Tang Qunzhi, Wu Haocheng, L. Zhigang, Lv Hongwei, Tang Ke
Aiming at the problem of water mist condensation on the fiber end face in a high-power fiber laser system, the most important factor causing this problem is that the traditional optical fiber connector does not have the mois-ture-proof sealing performance. The connector structure assembly and use process are analyzed in-depth, and the causes of the moisture-proof seal defects are pointed out. Through technological innovation and process improvement, a moisture-proof seal fiber connector is designed and completed. The principle and structure of the moisture-proof seal of the new connector are introduced. The main performances of the new connector are tested comprehensively, including immersion test, constant damp heat test, online application test. The experimental results show that the new connector has a better moisture-proof seal with IL less than 0.2 dB.
{"title":"Moisture-proof seal optical fiber connector","authors":"Zhang Xinli, Tang Qunzhi, Wu Haocheng, L. Zhigang, Lv Hongwei, Tang Ke","doi":"10.12086/OEE.2020.200067","DOIUrl":"https://doi.org/10.12086/OEE.2020.200067","url":null,"abstract":"Aiming at the problem of water mist condensation on the fiber end face in a high-power fiber laser system, the most important factor causing this problem is that the traditional optical fiber connector does not have the mois-ture-proof sealing performance. The connector structure assembly and use process are analyzed in-depth, and the causes of the moisture-proof seal defects are pointed out. Through technological innovation and process improvement, a moisture-proof seal fiber connector is designed and completed. The principle and structure of the moisture-proof seal of the new connector are introduced. The main performances of the new connector are tested comprehensively, including immersion test, constant damp heat test, online application test. The experimental results show that the new connector has a better moisture-proof seal with IL less than 0.2 dB.","PeriodicalId":39552,"journal":{"name":"Guangdian Gongcheng/Opto-Electronic Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86418098","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 : 2020-12-22DOI: 10.12086/OEE.2020.190669
Wang Ronggui, Wang Jing, Yang Juan, Xue Lixia
Existing works in person re-identification only considers extracting invariant feature representations from cross-view visible cameras, which ignores the imaging feature in infrared domain, such that there are few studies on visible-infrared relevant modality. Besides, most works distinguish two-views by often computing the similarity in feature maps from one single convolutional layer, which causes a weak performance of learning features. To handle the above problems, we design a feature pyramid random fusion network (FPRnet) that learns discriminative multiple semantic features by computing the similarities between multi-level convolutions when matching the person. FPRnet not only reduces the negative effect of bias in intra-modality, but also balances the heterogeneity gap between inter-modality, which focuses on an infrared image with very different visual properties. Meanwhile, our work integrates the advantages of learning local and global feature, which effectively solves the problems of visible-infrared person re-identification. Extensive experiments on the public SYSU-MM01 dataset from aspects of mAP and convergence speed, demonstrate the superiorities in our approach to the state-of-the-art methods. Furthermore, FPRnet also achieves competitive results with 32.12% mAP recognition rate and much faster convergence.
{"title":"Feature pyramid random fusion network for visible-infrared modality person re-identification","authors":"Wang Ronggui, Wang Jing, Yang Juan, Xue Lixia","doi":"10.12086/OEE.2020.190669","DOIUrl":"https://doi.org/10.12086/OEE.2020.190669","url":null,"abstract":"Existing works in person re-identification only considers extracting invariant feature representations from cross-view visible cameras, which ignores the imaging feature in infrared domain, such that there are few studies on visible-infrared relevant modality. Besides, most works distinguish two-views by often computing the similarity in feature maps from one single convolutional layer, which causes a weak performance of learning features. To handle the above problems, we design a feature pyramid random fusion network (FPRnet) that learns discriminative multiple semantic features by computing the similarities between multi-level convolutions when matching the person. FPRnet not only reduces the negative effect of bias in intra-modality, but also balances the heterogeneity gap between inter-modality, which focuses on an infrared image with very different visual properties. Meanwhile, our work integrates the advantages of learning local and global feature, which effectively solves the problems of visible-infrared person re-identification. Extensive experiments on the public SYSU-MM01 dataset from aspects of mAP and convergence speed, demonstrate the superiorities in our approach to the state-of-the-art methods. Furthermore, FPRnet also achieves competitive results with 32.12% mAP recognition rate and much faster convergence.","PeriodicalId":39552,"journal":{"name":"Guangdian Gongcheng/Opto-Electronic Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75245521","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 : 2020-11-20DOI: 10.12086/OEE.2020.190660
Wang Shuo, Guo Yong, Yang Lidong
Frequency modulated (FM) signal is extensively applied in sonar, radar, laser and emerging optical cross-research, its sparsity is a common basic issue for the sampling, denoising and compression of FM signal. This paper mainly studies the sparsity of FM signal in the fractional Fourier transform (FRFT) domain, and a maximum singular value method (MSVM) is proposed to estimate the compact FRFT domain of FM signal. This method uses the maximum singular value of amplitude spectrum of FM signal to measure the compact domain, and WOA is used to search the compact domain, which effectively improves the shortcomings of the existing methods. Compared with MNM and MACF, this method gives a sparser representation of FM signal in the FRFT domain, which has less number of significant amplitudes. Finally, the primary application of this method in the FM signal filtering is given.
{"title":"Research on sparsity of frequency modulated signal in fractional Fourier transform domain","authors":"Wang Shuo, Guo Yong, Yang Lidong","doi":"10.12086/OEE.2020.190660","DOIUrl":"https://doi.org/10.12086/OEE.2020.190660","url":null,"abstract":"Frequency modulated (FM) signal is extensively applied in sonar, radar, laser and emerging optical cross-research, its sparsity is a common basic issue for the sampling, denoising and compression of FM signal. This paper mainly studies the sparsity of FM signal in the fractional Fourier transform (FRFT) domain, and a maximum singular value method (MSVM) is proposed to estimate the compact FRFT domain of FM signal. This method uses the maximum singular value of amplitude spectrum of FM signal to measure the compact domain, and WOA is used to search the compact domain, which effectively improves the shortcomings of the existing methods. Compared with MNM and MACF, this method gives a sparser representation of FM signal in the FRFT domain, which has less number of significant amplitudes. Finally, the primary application of this method in the FM signal filtering is given.","PeriodicalId":39552,"journal":{"name":"Guangdian Gongcheng/Opto-Electronic Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74718074","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 : 2020-11-20DOI: 10.12086/OEE.2020.200040
Yao Han, S. Fan, H. Yiping, Wang Teng, Zeng Xianglong
We demonstrate a mode-locked Yb-doped fiber laser (YDFL) that enables fiber high-order mode (HOM) oscillation inside the ring cavity, by using a pair of mode selective couplers (MSCs) as an effective mode converter, the optical fiber HOM is obtained. The central wavelength of MSC is located at 1064 nm, which can achieve 80 nm mode conversion bandwidth and 94% high-order mode purity. A mode-locked pulsed fiber laser with a 3 dB spectral width of 7.4 nm, a pulse repetition frequency of 10.9 MHz, and a radio frequency signal-to-noise ratio of 55 dB is obtained, and the slope efficiency of the output power is 2.3%. These results show that the HOM can be directly oscillated by the cascaded MSCs in the fiber laser and participated in the mode-locking process to obtain a pulsed HOM laser.
{"title":"Mode-locked Yb-doped fiber laser based on mode coupler","authors":"Yao Han, S. Fan, H. Yiping, Wang Teng, Zeng Xianglong","doi":"10.12086/OEE.2020.200040","DOIUrl":"https://doi.org/10.12086/OEE.2020.200040","url":null,"abstract":"We demonstrate a mode-locked Yb-doped fiber laser (YDFL) that enables fiber high-order mode (HOM) oscillation inside the ring cavity, by using a pair of mode selective couplers (MSCs) as an effective mode converter, the optical fiber HOM is obtained. The central wavelength of MSC is located at 1064 nm, which can achieve 80 nm mode conversion bandwidth and 94% high-order mode purity. A mode-locked pulsed fiber laser with a 3 dB spectral width of 7.4 nm, a pulse repetition frequency of 10.9 MHz, and a radio frequency signal-to-noise ratio of 55 dB is obtained, and the slope efficiency of the output power is 2.3%. These results show that the HOM can be directly oscillated by the cascaded MSCs in the fiber laser and participated in the mode-locking process to obtain a pulsed HOM laser.","PeriodicalId":39552,"journal":{"name":"Guangdian Gongcheng/Opto-Electronic Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82821119","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 : 2020-11-20DOI: 10.12086/OEE.2020.190747
Gong Jiamin, Zhang Chen, Hao Qianwen, Z. Lihong, Wang Jie
In order to further improve the performance index of second-order Raman fiber amplifier, the main parameters of second-order RFA were analyzed. First, a structural model that can be controlled by optical switches and switched between two modes of traditional second-order and traditional first-order RFA is designed. It is proved through simulation that second-order RFA can increase the system gain and improve noise performance. The gain performance of first-order RFA is optimized. The optimization goal is to reduce the flatness. The particle swarm optimization algorithm is used to optimize the configuration of the wavelength and power of the pump light. After further structural improvement, a second-order RFA with a gain of 24.50 dB and a gain flatness of 0.98 dB were achieved in a 100 nm bandwidth. These results provide a reference for the design of second-order RFA with better performance in the future.
{"title":"Research on second-order Raman fiber amplifier based on particle swarm optimization","authors":"Gong Jiamin, Zhang Chen, Hao Qianwen, Z. Lihong, Wang Jie","doi":"10.12086/OEE.2020.190747","DOIUrl":"https://doi.org/10.12086/OEE.2020.190747","url":null,"abstract":"In order to further improve the performance index of second-order Raman fiber amplifier, the main parameters of second-order RFA were analyzed. First, a structural model that can be controlled by optical switches and switched between two modes of traditional second-order and traditional first-order RFA is designed. It is proved through simulation that second-order RFA can increase the system gain and improve noise performance. The gain performance of first-order RFA is optimized. The optimization goal is to reduce the flatness. The particle swarm optimization algorithm is used to optimize the configuration of the wavelength and power of the pump light. After further structural improvement, a second-order RFA with a gain of 24.50 dB and a gain flatness of 0.98 dB were achieved in a 100 nm bandwidth. These results provide a reference for the design of second-order RFA with better performance in the future.","PeriodicalId":39552,"journal":{"name":"Guangdian Gongcheng/Opto-Electronic Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79854450","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 : 2020-11-20DOI: 10.12086/OEE.2020.200030
Duan Chenghong, Chen Xiaokui, Luo Xiang-peng
Laser cleaning is a surface engineering technology which uses the high-energy characteristic of laser to remove attachments on the substrate. It has been gradually promoted in the fields of microelectronics, mould, building, aerospace, and so on. However, as a pillar industry of national economy, there are rare reports on the application of laser cleaning technology in petrochemical industry. In this paper, the cleaning needs and technology status of petrochemical industry are introduced, the existing technology cannot fully meet the new requirements of petrochemical industry development. The development of laser cleaning technology both at home and abroad in recent years is reviewed, the work has provided a theoretical basis and ideal reference for the application of laser cleaning in the petrochemical field. And the application occasions of laser cleaning technology in petrochemical field are analyzed. In addition, the specific research directions and application prospect are pointed out.
{"title":"Application prospect of laser cleaning in petrochemical field","authors":"Duan Chenghong, Chen Xiaokui, Luo Xiang-peng","doi":"10.12086/OEE.2020.200030","DOIUrl":"https://doi.org/10.12086/OEE.2020.200030","url":null,"abstract":"Laser cleaning is a surface engineering technology which uses the high-energy characteristic of laser to remove attachments on the substrate. It has been gradually promoted in the fields of microelectronics, mould, building, aerospace, and so on. However, as a pillar industry of national economy, there are rare reports on the application of laser cleaning technology in petrochemical industry. In this paper, the cleaning needs and technology status of petrochemical industry are introduced, the existing technology cannot fully meet the new requirements of petrochemical industry development. The development of laser cleaning technology both at home and abroad in recent years is reviewed, the work has provided a theoretical basis and ideal reference for the application of laser cleaning in the petrochemical field. And the application occasions of laser cleaning technology in petrochemical field are analyzed. In addition, the specific research directions and application prospect are pointed out.","PeriodicalId":39552,"journal":{"name":"Guangdian Gongcheng/Opto-Electronic Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85679350","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 : 2020-11-20DOI: 10.12086/OEE.2020.190628
Xue Lixia, Zhu Zhengfa, Wang Ronggui, Yang Juan
Person re-identification is significant but a challenging task in the computer visual retrieval, which has a wide range of application prospects. Background clutters, arbitrary human pose, and uncontrollable camera angle will greatly hinder person re-identification research. In order to extract more discerning person features, a network architecture based on multi-division attention is proposed in this paper. The network can learn the robust and dis-criminative person feature representation from the global image and different local images simultaneously, which can effectively improve the recognition of person re-identification tasks. In addition, a novel dual local attention network is designed in the local branch, which is composed of spatial attention and channel attention and can optimize the extraction of local features. Experimental results show that the mean average precision of the network on the Market-1501, DukeMTMC-reID, and CUHK03 datasets reaches 82.94%, 72.17%, and 71.76%, respectively.
{"title":"Person re-identification by multi-division attention","authors":"Xue Lixia, Zhu Zhengfa, Wang Ronggui, Yang Juan","doi":"10.12086/OEE.2020.190628","DOIUrl":"https://doi.org/10.12086/OEE.2020.190628","url":null,"abstract":"Person re-identification is significant but a challenging task in the computer visual retrieval, which has a wide range of application prospects. Background clutters, arbitrary human pose, and uncontrollable camera angle will greatly hinder person re-identification research. In order to extract more discerning person features, a network architecture based on multi-division attention is proposed in this paper. The network can learn the robust and dis-criminative person feature representation from the global image and different local images simultaneously, which can effectively improve the recognition of person re-identification tasks. In addition, a novel dual local attention network is designed in the local branch, which is composed of spatial attention and channel attention and can optimize the extraction of local features. Experimental results show that the mean average precision of the network on the Market-1501, DukeMTMC-reID, and CUHK03 datasets reaches 82.94%, 72.17%, and 71.76%, respectively.","PeriodicalId":39552,"journal":{"name":"Guangdian Gongcheng/Opto-Electronic Engineering","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86051830","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：