·1088· 智能系统学报 第16卷 characterizes a shadow boundary under the sun and 标阴影检测新方法[.光电子·激光，2018,29(12)： sky?[C]//Proceedings of the 2011 International Confer- 1317-1324. ence on Computer Vision.Barcelona,Spain,2011: ZHANG Degan,CHEN Chen,DONG Yue,et al.A new 898-905 method of moving object shadow detection based on ma- [2]FINLAYSON G D,HORDLEY S D,LU Cheng,et al.On chine learning[J].Journal of optoelectronicslaser,2018. the removal of shadows from images[J].IEEE transac- 2912):1317-1324 tions on pattern analysis and machine intelligence,2006, [11]HOSSEINZADEH S.SHAKERI M.ZHANG H.Fast 28(1):59-68. shadow detection from a single image using a patched [3]YU Xiaoming,LI Ge,YING Zhenqiang,et al.A new convolutional neural network[C]//2018 IEEE/RSJ Inter- shadow removal method using color-lines[C]//Interna- national Conference on Intelligent Robots and Systems tional Conference on Computer Analysis of Images and (IROS).Madrid.Spain.2018:1-5. Patterns.Ystad,Sweden,2017:307-319. [12]HU Xiaowei,JIANG Yitong,FU C W,et al.Mask- [4]林雨准，张保明，郭海涛，等.结合多尺度分割和形态学 ShadowGAN:learning to remove shadows from un- 运算的高分辨率遥感影像阴影检测).中国图象图形 paired data[C]//2019 IEEE/CVF International Confer- 学报，2018,23(8)：1263-1272 ence on Computer Vision (ICCV).Seoul,Korea(South). LIN Yuzhun,ZHANG Baoming,GUO Haitao,et al. 2019:2472-2481. Shadow detection from high resolution remote sensing [13]FAN Hui,HAN Meng,LI Jinjiang.Image shadow re- imagery based on multi-scale segmentation and morpho- moval using end-to-end deep convolutional neural networks logy operation[J].Journal of image and graphics,2018, [J].Applied sciences,2019,9(5):1009-1026. 23(8):1263-1272. [14]HU Xiaowei,FU C W.ZHU Lei,et al.Direction-aware [5]GUO Ruiqi,DAI Qieyun,HOIEM D.Single-image shad- spatial context features for shadow detection and removal ow detection and removal using paired regions[C]//CV- [J].IEEE transactions on pattern analysis and machine PR 2011.Colorado Springs,USA,2011:2033-2040. intelligence,2020,42(11):2795-2808 [6]FAN Xinyun,WU Wenjun,ZHANG Ling,et al.Shading- [15]GOODFELLOW I POUGET-ABADIE J,MIRZA M.etal. aware shadow detection and removal from a single im- Generative adversarial nets[C]//Proceedings of the 27th age[J].The visual computer,2020,36(10/11/12): International Conference on Neural Information Pro- 2175-2188. cessing Systems.Montreal,Canada,2014:2672- [7]焦俊男，石静.田庆久，等.多光谱影像的NDVI阴影影 2680. 响去除模型[.遥感学报，2020,241)：53-66 [16]张瑞倩，邵振峰，PORTNOV A,等.多尺度空洞卷积的 JIAO Junnan,SHI Jing,TIAN Qingjiu,et al.Researchon 无人机影像目标检测方法.武汉大学学报（信息科 multispectral-image-based NDVI shadow-effect-eliminat- 学版).2020.45(6)：895-903 ing model[J].Journal of remote sensing,2020,24(1): ZHANG Ruiqian,SHAO Zhenfeng,PORTNOV A,et al. 53-66. Multi-scale dilated convolutional neural network for ob- [8]ZHANG Yindan,CHEN Gang,Vukomanovi J,et al.Re- ject detection in UAV images[J].Journal of Wuhan Uni- current Shadow Attention Model(RSAM)for shadow re- versity (information science edition),2020,45(6): moval in high-resolution urban land-cover mapping[J]. 895-903. Remote sensing of environment,2020,247:111945 [17]颜铭靖，苏喜友.基于三维空洞卷积残差神经网络的 [9]吴文，万毅.基于低尺度细节恢复的单幅图像阴影去除 高光谱影像分类方法J].光学学报，2020,40(16) 方法U.电子学报，2020,48(7)：1293-1302. 1628002. WU Wen,WAN Yi.Single image shadow removal using YAN Mingjing,SU Xiyou.Hyperspectral image classi- low-scale detail recovering[J].Acta electronica sinica, fication based on three-dimensional dilated convolution- 2020,48(7):1293-1302. al residual neural network[J].Acta optica sinica,2020 [10]张德干，陈晨，董悦，等.一种基于机器学习的运动目 40(16):1628002characterizes a shadow boundary under the sun and sky?[C]//Proceedings of the 2011 International Confer￾ence on Computer Vision. Barcelona, Spain, 2011: 898−905. FINLAYSON G D, HORDLEY S D, LU Cheng, et al. On the removal of shadows from images[J]. IEEE transac￾tions on pattern analysis and machine intelligence, 2006, 28(1): 59–68. [2] YU Xiaoming, LI Ge, YING Zhenqiang, et al. A new shadow removal method using color-lines[C]//Interna￾tional Conference on Computer Analysis of Images and Patterns. Ystad, Sweden, 2017: 307−319. [3] 林雨准, 张保明, 郭海涛, 等. 结合多尺度分割和形态学 运算的高分辨率遥感影像阴影检测 [J]. 中国图象图形 学报, 2018, 23(8): 1263–1272. LIN Yuzhun, ZHANG Baoming, GUO Haitao, et al. Shadow detection from high resolution remote sensing imagery based on multi-scale segmentation and morpho￾logy operation[J]. Journal of image and graphics, 2018, 23(8): 1263–1272. [4] GUO Ruiqi, DAI Qieyun, HOIEM D. Single-image shad￾ow detection and removal using paired regions[C]//CV￾PR 2011. Colorado Springs, USA, 2011: 2033−2040. [5] FAN Xinyun, WU Wenjun, ZHANG Ling, et al. Shading￾aware shadow detection and removal from a single im￾age[J]. The visual computer, 2020, 36(10/11/12): 2175–2188. [6] 焦俊男, 石静, 田庆久, 等. 多光谱影像的 NDVI 阴影影 响去除模型 [J]. 遥感学报, 2020, 24(1): 53–66. JIAO Junnan, SHI Jing, TIAN Qingjiu, et al. Researchon multispectral-image-based NDVI shadow-effect-eliminat￾ing model[J]. Journal of remote sensing, 2020, 24(1): 53–66. [7] ZHANG Yindan, CHEN Gang, Vukomanovi J, et al. Re￾current Shadow Attention Model (RSAM) for shadow re￾moval in high-resolution urban land-cover mapping[J]. Remote sensing of environment, 2020, 247: 111945. [8] 吴文, 万毅. 基于低尺度细节恢复的单幅图像阴影去除 方法 [J]. 电子学报, 2020, 48(7): 1293–1302. WU Wen, WAN Yi. Single image shadow removal using low-scale detail recovering[J]. Acta electronica sinica, 2020, 48(7): 1293–1302. [9] [10] 张德干, 陈晨, 董悦, 等. 一种基于机器学习的运动目 标阴影检测新方法 [J]. 光电子·激光, 2018, 29(12): 1317–1324. ZHANG Degan, CHEN Chen, DONG Yue, et al. A new method of moving object shadow detection based on ma￾chine learning[J]. Journal of optoelectronics·laser, 2018, 29(12): 1317–1324. HOSSEINZADEH S, SHAKERI M, ZHANG H. Fast shadow detection from a single image using a patched convolutional neural network[C]//2018 IEEE/RSJ Inter￾national Conference on Intelligent Robots and Systems (IROS). Madrid, Spain, 2018: 1−5. [11] HU Xiaowei, JIANG Yitong, FU C W, et al. Mask￾ShadowGAN: learning to remove shadows from un￾paired data[C]//2019 IEEE/CVF International Confer￾ence on Computer Vision (ICCV). Seoul, Korea (South), 2019: 2472−2481. [12] FAN Hui, HAN Meng, LI Jinjiang. Image shadow re￾moval using end-to-end deep convolutional neural networks [J]. Applied sciences, 2019, 9(5): 1009–1026. [13] HU Xiaowei, FU C W, ZHU Lei, et al. Direction-aware spatial context features for shadow detection and removal [J]. IEEE transactions on pattern analysis and machine intelligence, 2020, 42(11): 2795–2808. [14] GOODFELLOW I, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial nets[C]//Proceedings of the 27th International Conference on Neural Information Pro￾cessing Systems. Montreal, Canada, 2014: 2672− 2680. [15] 张瑞倩, 邵振峰, PORTNOV A, 等. 多尺度空洞卷积的 无人机影像目标检测方法 [J]. 武汉大学学报(信息科 学版), 2020, 45(6): 895–903. ZHANG Ruiqian, SHAO Zhenfeng, PORTNOV A, et al. Multi-scale dilated convolutional neural network for ob￾ject detection in UAV images[J]. Journal of Wuhan Uni￾versity (information science edition), 2020, 45(6): 895–903. [16] 颜铭靖, 苏喜友. 基于三维空洞卷积残差神经网络的 高光谱影像分类方法 [J]. 光学学报, 2020, 40(16): 1628002. YAN Mingjing, SU Xiyou. Hyperspectral image classi￾fication based on three-dimensional dilated convolution￾al residual neural network[J]. Acta optica sinica, 2020, 40(16): 1628002. [17] ·1088· 智 能 系 统 学 报 第 16 卷