基于支持向量机的遥感影像辐射质量评价方法

当代摄影测量与遥感技术的迅速发展,已经步入大数据时代,如何对获得的海量数字影像的辐射质量进行评价是一个值得重视的问题。本文从信息量、清晰度、灰度分布3个方面选择了10个评价指标作为影像特征,利用支持向量机监督学习的方法对以资源三号为例的遥感影像的辐射质量进行评价及结果分析。试验结果表明,本文方法得到的评价结果与人工评价结果较为一致,准确度较高,并且自动化程度高,可应用于遥感影像的辐射质量评价。

Radiation of Remote Sensing Image Quality Assessment Based on Support Vector Machine ( SVM) Method

The rapid development of the contemporary photogrammetry and remote sensing technology , has been into the era of big da-ta, how to obtain the huge amounts of digital image radiation quality evaluation is a serious problem .In three respects:information, clarity, grayscale distribution ten evaluation index as the image characteristics , the method of using support vector machine (SVM) learning supervision resources No .3 as an example to evaluate the radiation quality of remote sensing image and the result analysis . Experimental results show that the method of the evaluation results with artificial evaluation results are consistent , high accuracy , and high degree of automation , suitable for application in remote sensing image quality evaluation of radiation .