基于光学遥感图像的土地利用空间格局识别  

Spatial Pattern Recognition of Land Use Based on Optical Remote Sensing Images

作　　者：李彦粉 Li Yanfen(Beijing Vitspace Technology INC.)

机构地区：[1]北京威特空间科技有限公司,北京市102600

出　　处：《勘察科学技术》2025年第1期68-74,共7页Site Investigation Science and Technology

摘　　要：土地利用空间在格局上的分界差异很小,当前土地利用空间格局识别过程中主要使用以电磁波为主的普通遥感图像,受困于采集的土地空间遥感特征尺度差异大,几何偏差突变区域多,导致土地利用空间格局识别效果差,该文提出了基于光学遥感图像的土地利用空间格局识别方法。该方法利用无人机采集遥感图像,设计一种根据基准图像和校正图像特征对应关系的图像拼接技术,以克服当前大差异土地光学遥感图像几何校正的弊端;设计了一种暗通道和双边滤波相结合的方法,对校正后的图像进行增强处理,该方法在滤波过程中考虑了像素之间的空间距离和灰度相似性,双边滤波能够降低图像噪声并保持图像细节,同时避免了传统高斯滤波对边缘信息的模糊处理;通过融合多尺度与主方向因子的灰度共生矩阵纹理特征,完成土地利用空间格局的识别。实验结果表明,该方法对土地利用空间格局识别准确率高、效率高且效果好。The difference in the pattern of boundary of land use space is very small.Currently,in the process of identifying land use spatial patterns,ordinary remote sensing images dominated by electromagnetic waves.However,due to the difficulty of large differences in the scale of land spatial remote sensing feature collection and multiple areas of sudden geometric deviation mutation,leading to poor land use spatial pattern identification,a land use spatial pattern recognition method based on optical remote sensing images is proposed in this paper.Using drones to collect remote sensing images,this method design an image splicing technique based on the correspondence between the features of the reference image and the corrected image,in order to overcome the drawbacks of current geometric correction in optical remote sensing images of large-difference land.It designs a method that combines dark channel and bilateral filtering to enhance the corrected image,taking into account the spatial distance and grayscale similarity between pixels during the filtering process.Bilateral filtering can reduce image noise and maintain image details,while avoiding the fuzzy processing of edge information by traditional Gaussian filtering.By fusing the texture features of the grayscale co-occurrence matrix with multi-scale and principal directional factors,the recognition of land use spatial patterns is achieved.The experimental results show that the proposed method has high accuracy,efficiency,and effectiveness in identifying land use spatial patterns.

关 键 词：光学遥感图像 土地利用空间 格局识别 几何校正 图像增强 

分 类 号：P237[天文地球—摄影测量与遥感]