遥感图像去云算法研究
|
摘要
由于在遥感成像的区域中存在云的影响,从遥感图像中无法获取有云区域中的详细信息,因此对遥感图像的去云技术研究成为图像增强领域的研究热点。文章基于小波变换理论的遥感图像增强,提出一种基于中值滤波和小波分析结合运用于遥感图像增强的新算法。在计算机自动实现和用户交互实现时,将小波工具引入到云层提取、处理的过程中来,实现从单幅图像中去除薄云,改善效果而不增加其他副作用。对比遥感图像原图的直方图和增强后的直方图,经过新算法处理后图像均值变低,平均灰度降低。结果表明:该算法在去噪的同时能保留大量的图像边缘细节等重要信息,具有非常好的去噪效果。
Because of the influence of cloud in the remote sensing image region, it is impossible to obtain the detailed information of the cloud region from the remote sensing image, so the research on cloud removal technology of remote sensing image has become a research hotspot in the field of image enhancement. Based on wavelet transform theory, a new algorithm for remote sensing image enhancement based on median filtering and wavelet analysis is proposed in this paper. When the interaction between computer and user is implemented, wavelet tools are introduced into the process of cloud extraction and processing to remove thin clouds from a single image and improve the effect without increasing other side effects. Compared with the original histogram of remote sensing image and the enhanced histogram, the average value of image is lower and the average gray level is lower after the new algorithm processing. The results show that the algorithm can retain a lot of important information such as image edge details while denoising,and has a very good denoising effect.
Because of the influence of cloud in the remote sensing image region, it is impossible to obtain the detailed information of the cloud region from the remote sensing image, so the research on cloud removal technology of remote sensing image has become a research hotspot in the field of image enhancement. Based on wavelet transform theory, a new algorithm for remote sensing image enhancement based on median filtering and wavelet analysis is proposed in this paper. When the interaction between computer and user is implemented, wavelet tools are introduced into the process of cloud extraction and processing to remove thin clouds from a single image and improve the effect without increasing other side effects. Compared with the original histogram of remote sensing image and the enhanced histogram, the average value of image is lower and the average gray level is lower after the new algorithm processing. The results show that the algorithm can retain a lot of important information such as image edge details while denoising,and has a very good denoising effect.
引文
[1]毕晓君,潘铁文.基于改进的教与学优化算法的图像增强[J].哈尔滨工程大学学报,2016,37(12):1716-1721.
[2]苏娟,李冰,王延钊.结合PCNN分割和模糊集理论的红外图像增强[J].光学学报,2016(9):82-90.
[3]代书博,徐伟,朴永杰,等.基于暗原色先验的遥感图像去雾方法[J].光学学报,2017(3):341-347.
[4]张志强,张新长,辛秦川,等.结合像元级和目标级的高分辨率遥感影像建筑物变化检测[J].测绘学报,2018(1):102-112.
[5]吴田军,夏列钢,吴炜,等.土地执法监察中的高分辨率遥感及变化检测技术[J].地球信息科学学报,2016,18(7):962-968.
[6]Zhou L,Cao G,Li Y,et al.Change Detection Based on Conditional Random Field With Region Connection Constraints in High-Resolution Remote Sensing Images[J].IEEE Journal of Selected Topics in Applied Earth Observations&Remote Sensing,2017,9(8):3478-3488.
[7]胡根生,查慧敏,梁栋,等.结合分类与迁移学习的薄云覆盖遥感图像地物信息恢复[J].电子学报,2017(12):2855-2862.
[8]Cohen E,Picard R H,Crabtree P N.Spectral-overlap approach to multiframe superresolution image reconstruction[J].Applied Optics,2016,55(15):3978.
[9]晏建洋,吴建星.基于小波变换的微震信号去噪研究[J].科技通报,2016,32(3):185-188.
[10]秦冬冬,陈志军,闫学勤.多层阈值函数下的小波图像去噪[J].计算机工程,2017,43(6):202-206.
[11]蔡猛.基于小波变换的多传感器最优信息融合[J].电光与控制,2013,20(12):97-100.
[12]余竹,夏禾,Goicolea,等.基于小波包能量曲率差法的桥梁损伤识别试验研究[J].振动与冲击,2013,32(5):20-25.
[13]王金亮,曾浩,王艳英,等.基于小波分析的TM遥感图像超分辨率重建[J].遥感技术与应用,2016,31(3):476-480.
[14]肖书敏,闫云聚,姜波澜.基于小波神经网络方法的桥梁结构损伤识别研究[J].应用数学和力学,2016,37(2):149-159.
[15]胡根生,查慧敏,梁栋,等.结合分类与迁移学习的薄云覆盖遥感图像地物信息恢复[J].电子学报,2017,45(12):2855-2862.
[16]Song Q,Ma L,Cao J,et al.Image Denoising Based on Mean Filter and Wavelet Transform[C]//International Conference on Advanced Information Technology and Sensor Application.IEEE,2016:39-42.
[17]徐少平,张兴强,姜尹楠,等.局部均值噪声估计的盲3维滤波降噪算法[J].中国图象图形学报,2017,22(4):422-434.
[18]Xu C.Detecting periodic oscillations in astronomy data:revisiting wavelet analysis with coloured and white noise[J].Monthly Notices of the Royal Astronomical Society,2017,466(4):3827-3833.
[2]苏娟,李冰,王延钊.结合PCNN分割和模糊集理论的红外图像增强[J].光学学报,2016(9):82-90.
[3]代书博,徐伟,朴永杰,等.基于暗原色先验的遥感图像去雾方法[J].光学学报,2017(3):341-347.
[4]张志强,张新长,辛秦川,等.结合像元级和目标级的高分辨率遥感影像建筑物变化检测[J].测绘学报,2018(1):102-112.
[5]吴田军,夏列钢,吴炜,等.土地执法监察中的高分辨率遥感及变化检测技术[J].地球信息科学学报,2016,18(7):962-968.
[6]Zhou L,Cao G,Li Y,et al.Change Detection Based on Conditional Random Field With Region Connection Constraints in High-Resolution Remote Sensing Images[J].IEEE Journal of Selected Topics in Applied Earth Observations&Remote Sensing,2017,9(8):3478-3488.
[7]胡根生,查慧敏,梁栋,等.结合分类与迁移学习的薄云覆盖遥感图像地物信息恢复[J].电子学报,2017(12):2855-2862.
[8]Cohen E,Picard R H,Crabtree P N.Spectral-overlap approach to multiframe superresolution image reconstruction[J].Applied Optics,2016,55(15):3978.
[9]晏建洋,吴建星.基于小波变换的微震信号去噪研究[J].科技通报,2016,32(3):185-188.
[10]秦冬冬,陈志军,闫学勤.多层阈值函数下的小波图像去噪[J].计算机工程,2017,43(6):202-206.
[11]蔡猛.基于小波变换的多传感器最优信息融合[J].电光与控制,2013,20(12):97-100.
[12]余竹,夏禾,Goicolea,等.基于小波包能量曲率差法的桥梁损伤识别试验研究[J].振动与冲击,2013,32(5):20-25.
[13]王金亮,曾浩,王艳英,等.基于小波分析的TM遥感图像超分辨率重建[J].遥感技术与应用,2016,31(3):476-480.
[14]肖书敏,闫云聚,姜波澜.基于小波神经网络方法的桥梁结构损伤识别研究[J].应用数学和力学,2016,37(2):149-159.
[15]胡根生,查慧敏,梁栋,等.结合分类与迁移学习的薄云覆盖遥感图像地物信息恢复[J].电子学报,2017,45(12):2855-2862.
[16]Song Q,Ma L,Cao J,et al.Image Denoising Based on Mean Filter and Wavelet Transform[C]//International Conference on Advanced Information Technology and Sensor Application.IEEE,2016:39-42.
[17]徐少平,张兴强,姜尹楠,等.局部均值噪声估计的盲3维滤波降噪算法[J].中国图象图形学报,2017,22(4):422-434.
[18]Xu C.Detecting periodic oscillations in astronomy data:revisiting wavelet analysis with coloured and white noise[J].Monthly Notices of the Royal Astronomical Society,2017,466(4):3827-3833.