面向对象的干旱区盐渍地信息提取方法研究

随着高分辨率遥感影像的广泛应用,面向对象的高分辨率遥感影像信息提取技术得到了迅速发展。本文针对高分辨率遥感影像的盐渍地信息提取,探索了面向对象的信息提取技术运用在盐渍地信息提取方面的最优参数选择和规则建立等关键技术,并且将从ALOS影像中提取盐渍地信息的结果与传统分类方法的提取结果进行对比分析。结果表明,利用面向对象方法的盐渍地信息提取总体精度为89.38%,较传统方法提高8.24%,Kappa系数为0.88,较传统方法提高0.08。该方法能实现较为精确地提取盐渍地信息,在高分辨率遥感影像盐渍地信息提取中具有一定的优势。

Object-Oriented Information Extraction Method for Soil Salinization in Arid Area

Soil salinization is a major environmental issue in the world, and it is more serious in arid and semi-arid area. The remote sensing technology is widely used to detect and monitor soil salinizaion to timely provide the information about properties, spatial distribution and extent of soil salinization. The method of object-oriented information extraction from high resolution images has been developed rapidly with the wide use of high resolution remote sensing images. In this paper, the object-oriented classification method was used to extract soil salinization information from the ALOS images, and the result was compared with the traditional maximum likelihood classification method. The results show that the overall accuracy of this object-oriented method was 89.38% with a kappa coefficient value of 0.88. Compared with the traditional method, the overall accuracy and kappa coefficient value increased 8.24% and 0.08, respectively; The object-oriented method is superior to traditional method in high resolution remote sensing information extraction.