论文部分内容阅读
目前在基于内容的遥感图像检索中,存在图像内容描述不精确,图像查询精度低以及图像查询结果排序错误的问题.就此问题,提出了一种基于遗传算法优化和支持向量机(SVM)的遥感图像检索方法.该方法首先利用遗传算法优化遥感图像特征组合参数,得到对图像内容描述精确度较高的遥感图像检索模型,然后结合优化后的图像相似度计算模型对传统的SVM相关反馈算法进行改进,将遥感图像的视觉特征信息考虑到相关反馈的图像相似度计算中,并根据用户的反馈信息对图像的相似度计算模型进行修正,从而提高遥感图像查询的精度.实验结果表明,该方法优化了图像查询的排序结果,提高了遥感图像的检索精度,有效地减少了图像检索的反馈次数.
At present, content-based remote sensing image retrieval has some problems, such as inaccurate description of image content, low precision of image query, and wrong sorting of image query results.In this paper, a remote sensing based on genetic algorithm optimization and support vector machine (SVM) Image retrieval method.Firstly, genetic algorithm was used to optimize remote sensing image feature combination parameters to get the remote sensing image retrieval model with high accuracy of description of the image content. Then the traditional SVM correlation feedback algorithm was combined with the optimized image similarity calculation model In this paper, the visual characteristic information of remote sensing image is taken into account in the similarity calculation of the relevant feedback, and the similarity calculation model of the image is modified according to the user’s feedback information so as to improve the precision of the remote sensing image query. Experimental results show that this method The sorting result of the image query is optimized, the retrieval precision of the remote sensing image is improved, and the feedback times of the image retrieval are reduced effectively.