摘要
大数据量遥感图像薄云的存在影响图像的清晰度,受到薄云分布不均匀性和随机性影响,其采样存在信息不完整和噪声干扰问题,使得特征不完整,影响图像透射率,导致图像薄云特征分析不准确,薄云去除效果差;为此,提出基于分块压缩感知的大数据量遥感图像薄云去除方法;先定义像素感知对象的分块矩阵,基于分块压缩感知算法计算采样峰值的信噪比参量,实现大数据量遥感图像采样,以解决采样信息不完整,存在噪声干扰的问题;然后利用所得采样结果,求解图像的空间特征、灰度剖面图特征与频率特征,完成大数据量遥感图像薄云特征分析,提升特征分析效果;最后参考优化去除因子与导向滤波优化透射率,改进大气光值,实现对薄云去除参量的分波段迭代,完成遥感图像薄云去除设计;实验结果表明,应用所述方法能够缩小薄云覆盖区域与其边缘区域之间的像素差值,从而使薄云区域的色温水平更趋近于整幅图像的色温均值;经过处理,去除薄云后的图像纹理清晰度达到了95%,有效提高了遥感图像的清晰度,应用效果良好。
The existence of thin cloud in large data remote sensing images affects the clarity of images,which is affected by the uneven distribution uniformity and randomness of thin cloud.Its sampling has the characteristics of incomplete information and noise interference,which makes the features incomplete and affects the transmittance of images,leading to inaccurate analysis of thin cloud features in images and poor removal effect of thin cloud.Therefore,a thin cloud removal method for large data remote sensing images based on block compression perception is proposed.Firstly,define the block matrix of the pixel perception object,and calculate the signal-to-noise ratio parameter of the sampling peak based on the block compression sensing algorithm to realize the remote sensing image sampling of large data amount,so as to solve incomplete sampling information and noise interference.Then,the obtained sampling results are used to solve the spatial features,grayscale profile features and frequency features of the images,complete the thin cloud feature analysis for a amount of large data remote sensing images,and improve the effectiveness of feature analysis.Finally,The reference optimization removal factor and guide filter are used to optimize the transmission rate,improve the atmospheric light value,realize the subband iteration of thin cloud removal parameters,and achieve the remote sensing image thin cloud removal design.Experimental results show that the proposed method can shrink the pixel difference between the thin cloud coverage area and its edge area,so that the color temperature of the thin cloud area is closer to the average color temperature of the whole image.After removing the thin cloud,the clarity of image texture reaches up to 95%,which effectively improves the clarity of remote sensing images,with a good application effect.
作者
李旸园
LI Yangyuan(Electronic Information Engineering,Xi'an Siyuan University School,Xi'an 710038,China)
出处
《计算机测量与控制》
2025年第8期293-300,共8页
Computer Measurement & Control
基金
西安思源学院校级重点项目(XASYZD-B2202)。
关键词
分块压缩感知
遥感图像
薄云去除
信噪比
去除因子
大气光值
block compression perception
remote sensing images
thin cloud removal
signal to noise ratio
remove factors
atmospheric light value
作者简介
李旸园(1980-),男,博士,副教授。
