【摘要】 针对人工变化信息提取效率低下而变化信息自动检测容易出现错提和漏提问题,以南京市浦口区为研究区,选用北京二号影像为数据源,基于光谱特征和纹理特征相结合的变化检测方法,选取优化土壤调节植被指数进行建设用地信息提取;提出综合特征指数对结果进行优化,通过纹理分析剔除伪变化信息,并获取规则的图斑边界。结果表明,该方法建设用地变化检测精度较高,通过人机交互对建设用地变化检测结果进行适当编辑完善基本可满足实际应用要求,具有较高的实用价值。更多还原

【Abstract】 The efficiency of manual extraction of change information is low.Automatic detection method is prone to be confused,which will lead to the error and omission extraction of change detection information.In this paper,a method of land change detection based on fusion of multiple features is proposed.This paper takes Pukou district of Nanjing as the research area.The BJ-2 image is selected as the data source.Optimization of soil adjusted vegetation index(OSAVI)is used to classify the information,comprehensive characteristic index(CCI)is put forward to optimize the classification results,and entropy and correlation coefficient are selected to analyze the texture characteristic.The results show that the method has high precision in land use change detection,and has high practical value in land use dynamic monitoring.更多还原