论文部分内容阅读
对遥感影像的计算机图像处理与识别技术是现阶段遥感技术研究的热点内容,本文对新疆渭干–库车河流域不同分辨率的全色影像与多光谱影像融合并构建多源空间数据库,再通过与面向对象分类法相结合,进行图像分割、特征提取后将结果取值输入Matlab软件,编制相应的算法程序,构建关联规则分类器,探索分类规则并进行遥感影像的分类.同样基于Matlab软件平台,根据最大似然分类算法对同一研究区影像进行分类,并对分类结果进行精度检验.结果表明:相较传统分类方法,关联规则方法拓宽了传统逻辑推理分类方法中获取信息的途径,使分类更加智能化,提高并增强了影像分类速度与算法可靠性,分类精度得到显著提高,Kapp系数为0.83.因此,关联规则方法更适用于影像分类,可以作为研究区有效信息提取的支持手段.
The computer image processing and recognition technology of remote sensing images are the hot topics in the research of remote sensing technology at present. In this paper, the multi-source spatial database is integrated with the panchromatic and multi-spectral images of different resolutions in the Weigan-Kuqa River basin of Xinjiang Through the combination with object-oriented classification, the image segmentation is carried out. After the feature extraction, the result is input into the Matlab software, the corresponding algorithm program is compiled, the association rule classifier is constructed, the classification rules are explored and the remote sensing images are classified. , Classifying the images of the same study area according to the maximum likelihood classification algorithm and checking the accuracy of the classification results.The results show that compared with the traditional classification methods, the association rules method broadens the access to information in the traditional logical reasoning classification method, Therefore, the association rule method is more suitable for image classification, which can be used as the support method for effective information extraction in the study area.