多时相遥感影像阴影角度精确校正仿真
|
摘要
为了增强遥感影像对实际拍摄区域的还原效果,提高合成影像信息的利用价值,针对当前影像阴影角度校正方法中存在的阴影区域提取不准确、补偿效果较差、角度校正过程所需时间长等问题,提出基于灰度补偿的多时相遥感影像阴影角度精确校正方法。分别计算遥感影像阴影区域的色调差值、蓝色通道与绿色通道的差值及亮度与饱和度差值,结合D-S证据理论将各差值结果融合作为颜色特征提取影像中的阴影区域。采用灰度线性变换算法对得到的阴影区域进行灰度补偿,并进行高灰度噪点滤除,实现影像阴影区域的校正与边缘平滑处理。计算阴影校正后的图像间差值与初始图像间差值,结合遥感影像设备的轨道运行参数构建阴影角度校正模型,利用模型完成影像阴影角度的精确校正。实验结果表明,所提方法阴影角度校正结果更接近真实值,校正耗时更短,具有较好的适用性。
This article presents an accurate correction method for shadow angle of multi-temporal remote sensing image based on gray compensation. Respectively, we calculated the hue difference, the difference between the blue channel and the green channel, and the difference between the brightness and the saturation degree in the shadow region of remote sensing image. Combined with D-S evidence theory, all difference results were mixed together as the color feature to extract the shadow region in image. Then, we used grayscale linear transformation algorithm to perform grayscale compensation on the obtained shadow area and conduct high grayscale noise filtering to achieve the correction of shadow area and edge smoothing processing. In addition, we calculated the difference between images after shadow correction and the difference between initial images. In combination with the orbital motion parameter of remote sensing imaging equipment, we built the shadow angle correction model, and then used the model to complete the accurate correction of shadow shading angle. According to simulation results, we can see that the shadow angle correction of proposed method is closer to actual value. Meanwhile, the correction time is shorter, which has better applicability.
This article presents an accurate correction method for shadow angle of multi-temporal remote sensing image based on gray compensation. Respectively, we calculated the hue difference, the difference between the blue channel and the green channel, and the difference between the brightness and the saturation degree in the shadow region of remote sensing image. Combined with D-S evidence theory, all difference results were mixed together as the color feature to extract the shadow region in image. Then, we used grayscale linear transformation algorithm to perform grayscale compensation on the obtained shadow area and conduct high grayscale noise filtering to achieve the correction of shadow area and edge smoothing processing. In addition, we calculated the difference between images after shadow correction and the difference between initial images. In combination with the orbital motion parameter of remote sensing imaging equipment, we built the shadow angle correction model, and then used the model to complete the accurate correction of shadow shading angle. According to simulation results, we can see that the shadow angle correction of proposed method is closer to actual value. Meanwhile, the correction time is shorter, which has better applicability.
引文
[1] 陈建珍,何超,寇卫利. 高山峡谷地区遥感图像地形校正[J]. 山地学报, 2016,34(5): 623-631.
[2] 邱春霞,等. 一种遥感影像反立体纠正方法研究[J]. 西安科技大学学报, 2016,36(3): 422-427.
[3] 刘时城,等. 不同地形校正方法对刺槐林遥感提取的影响[J]. 北京林业大学学报, 2017,39(5):25-33.
[4] 王相海,等. 基于零树小波的交通视频车辆运动阴影滤除方法[J]. 模式识别与人工智能, 2016,29(12):1104-113.
[5] 马超,等. 卫星遥感图像的几何校正优化研究仿真[J]. 计算机仿真, 2017,34(3):38-41.
[6] 温兆飞,等. 辐射特征支持下的城市高分影像阴影校正[J]. 遥感学报, 2016,20(1): 138-148.
[7] 边金虎,等. 基于MODISNDVI的LandsatTM影像地形阴影区光谱信息恢复方法研究[J]. 遥感技术与应用, 2016,31(1): 12-22.
[8] 王冰,等. QuickBird影像城市阴影信息的提取与消除[J]. 地球信息科学学报, 2016,18(2):255-262.
[9] 杨猛, 等. 集成特征分量的阴影多尺度分割方法[J]. 遥感信息, 2017,32(1):109-114.
[10] 武丹,刘涛,杨树文. 资源三号卫星高分影像的城市建筑物阴影提取[J]. 测绘科学, 2017,42(6):190-195.
[2] 邱春霞,等. 一种遥感影像反立体纠正方法研究[J]. 西安科技大学学报, 2016,36(3): 422-427.
[3] 刘时城,等. 不同地形校正方法对刺槐林遥感提取的影响[J]. 北京林业大学学报, 2017,39(5):25-33.
[4] 王相海,等. 基于零树小波的交通视频车辆运动阴影滤除方法[J]. 模式识别与人工智能, 2016,29(12):1104-113.
[5] 马超,等. 卫星遥感图像的几何校正优化研究仿真[J]. 计算机仿真, 2017,34(3):38-41.
[6] 温兆飞,等. 辐射特征支持下的城市高分影像阴影校正[J]. 遥感学报, 2016,20(1): 138-148.
[7] 边金虎,等. 基于MODISNDVI的LandsatTM影像地形阴影区光谱信息恢复方法研究[J]. 遥感技术与应用, 2016,31(1): 12-22.
[8] 王冰,等. QuickBird影像城市阴影信息的提取与消除[J]. 地球信息科学学报, 2016,18(2):255-262.
[9] 杨猛, 等. 集成特征分量的阴影多尺度分割方法[J]. 遥感信息, 2017,32(1):109-114.
[10] 武丹,刘涛,杨树文. 资源三号卫星高分影像的城市建筑物阴影提取[J]. 测绘科学, 2017,42(6):190-195.