论文部分内容阅读
采用面向对象遥感影像分类方法,进行了高空间分辨率遥感影像信息提取试验,分析了其与基于像元方法的信息提取结果的差异,试验研究表明,在目视效果上,传统方法的分类结果图中椒盐现象非常明显,而面向对象方法可以有效地避免椒盐现象;在分类精度上,面向对象方法分类结果的总体精度、Kappa系数、生产者精度、用户精度、Hellden精度和Short精度均明显高于传统方法,各类地物提取效果显著提高。面向对象方法在高空间分辨率遥感影像信息提取中具有明显的优势。
Object-oriented remote sensing image classification method was used to extract high spatial resolution remote sensing image information, and the difference between the method and the pixel-based information extraction was analyzed. The experimental results show that in the visual effect, the traditional method classification results The salt and pepper phenomenon in the figure is very obvious, while the object-oriented method can effectively avoid the salt and pepper phenomenon. The overall accuracy, Kappa coefficient, producer accuracy, user accuracy, Hellden precision and Short accuracy of the object-oriented method are obviously higher In the traditional method, the extraction effect of all kinds of ground objects is remarkably improved. Object-oriented method has obvious advantages in high spatial resolution remote sensing image information extraction.