激光与光电子学进展Pub Date : 2023-01-01DOI: 10.3788/lop222634
何哲 He Zhe, 陶于祥 Tao Yuxiang, 罗小波 Luo Xiaobo, 徐浩 Xu Hao
{"title":"基于改进U-Net的遥感图像道路提取","authors":"何哲 He Zhe, 陶于祥 Tao Yuxiang, 罗小波 Luo Xiaobo, 徐浩 Xu Hao","doi":"10.3788/lop222634","DOIUrl":"https://doi.org/10.3788/lop222634","url":null,"abstract":"从遥感图像中提取的道路信息在城市规划、交通管理等领域具有极其巨大的价值，但由于背景复杂、障碍物遮挡以及大量类似的非道路区域的干扰，目前从遥感图像中提取高质量道路信息仍受到限制。为此，提出一种基于混合尺度注意力U-Net的遥感图像道路提取方法HSA-UNet。首先，在编码网络使用由残差结构和注意力特征融合机制构成的注意力残差学习单元，提高对全局特征和局部特征的提取能力；其次，由于道路具有跨度大、狭窄且连续分布等特点，在桥接网络加入注意力增强的空洞空间金字塔池化模块来增强不同尺度道路特征提取能力。在Massachusetts道路数据集上进行的实验结果表明，在F1、交并比等评价指标上，HSA-UNet明显优于D-LinkNet、DeepLabV3+等语义分割网络。","PeriodicalId":51502,"journal":{"name":"激光与光电子学进展","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135152246","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
激光与光电子学进展Pub Date : 2023-01-01DOI: 10.3788/lop222740
邓超 Deng Chao, 高超 Gao Chao, 王晓茜 Wang Xiaoqian, 姚治海 Yao Zhihai
{"title":"基于图像融合的优化重构计算鬼成像","authors":"邓超 Deng Chao, 高超 Gao Chao, 王晓茜 Wang Xiaoqian, 姚治海 Yao Zhihai","doi":"10.3788/lop222740","DOIUrl":"https://doi.org/10.3788/lop222740","url":null,"abstract":"针对计算鬼成像采样过程中由于外界影响而出现的物体信息缺失问题，将小波分析应用于计算鬼成像系统中，同时在此系统中引入图像融合方法，构建一种基于图像融合的计算鬼成像系统，对此问题进行了仿真模拟与实验研究。并对结果进行了分析，结果显示此系统可以有效恢复待测物体信息、提高成像质量，相较于传统的计算鬼成像系统更加适用于实际应用环境。","PeriodicalId":51502,"journal":{"name":"激光与光电子学进展","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135662127","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
激光与光电子学进展Pub Date : 2023-01-01DOI: 10.3788/lop222356
王明军 Wang Mingjun, 屠思凯 Tu Sikai
{"title":"基于反向传播神经网络盲均衡算法的水下涡旋光复用系统性能研究","authors":"王明军 Wang Mingjun, 屠思凯 Tu Sikai","doi":"10.3788/lop222356","DOIUrl":"https://doi.org/10.3788/lop222356","url":null,"abstract":"在水下信道中使用涡旋光复用通信技术可以有效提高通信系统的信道容量，然而海洋湍流会引起涡旋光束的模态间串扰，造成通信系统的性能下降。为了缓解模态串扰问题，本文引入基于反向传播（BP）神经网络的盲均衡算法，采取4路涡旋光进行复用传输，使用随机相位屏法模拟海洋湍流，仿真分析了系统在加入BP盲均衡算法后，改变海洋湍流强度、传输距离以及涡旋光复用模式等因素下，系统误码率的改善情况。仿真结果表明，利用BP神经网络的盲均衡算法能够有效降低海洋湍流对系统误码率的影响，且在复用模式间隔选为2时，系统性能改善明显。","PeriodicalId":51502,"journal":{"name":"激光与光电子学进展","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135784415","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","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
求助内容：
应助结果提醒方式：