论文部分内容阅读
旨在综合利用特征提取、多种分类器、图像融合、相关判决等方法,实现对SAR遥感图像中农田种植区域的精确识别与检测。首先论述了相关研究与应用现状,然后分析了SAR图像农田种植区域的特征提取、多种分类器训练与检测的效果,及采用基于PCA的SAR图像融合、多种检测结果的相关,最后综合以上提出了一种基于多分类器集成学习的SAR图像农田区域识别与检测方法。通过实验验证了文章所提出方法的有效性。
The purpose of this paper is to realize the accurate identification and detection of farmland planting areas in SAR remote sensing images by comprehensively utilizing the methods of feature extraction, multiple classifiers, image fusion and related judgments. Firstly, the status quo of related research and application is discussed. Then the feature extraction of cropland in the field of SAR images, the training and detection of multiple classifiers are analyzed, and the correlation of SAR images based on PCA and multiple detection results is analyzed. Finally, A method of agricultural land area identification and detection based on multi-classifier integrated learning is proposed. The validity of the proposed method is verified through experiments.