Feature extraction and matching method based on multimodal image of airborne photoelectric system

The invention belongs to the technical field of airborne photoelectric reconnaissance and situation awareness, and discloses a feature extraction and matching method based on a multi-modal image of an airborne photoelectric system, which comprises the following steps: S1, extracting features of the multi-modal image; s2, adaptively selecting a matching template; s3, performing kernel correlation matching based on a template; and S4, removing mismatching. According to the method, the feature map is generated after feature extraction is carried out on the multi-modal image of the airborne photoelectric system, matching of a large number of feature points is carried out based on the feature map, mismatching is eliminated, correct matching point pairs are obtained, finally correct matching of the multi-modal image is achieved, and the capacity under multiple task scenes can be improved. 本发明属于机载光电侦察、态势感知技术领域，公开了一种基于机载光电系统多模态图像的特征提取与匹配方法，其包括以下步骤：S1：多模态图像特征提取；S2：匹配模板自适应选择；S3：基于模板的核相关匹配；S4：误匹配剔除。本发明在对机载光电系统的多模态图像进行特征