Template matching is one of the fundamental methods for image recognition and target localization,making the selection of a more efficient template matching approach particularly important.First,the methods for extracting image features are discussed,including statistical feature extraction,Hu invariant moment feature extraction,and texture feature extraction.Then,the implementation process of template matching is detailed,including feature point extraction and similarity calculation.Finally,an analysis of the template matching results based on different image feature extraction methods is presented,the comparison of simulation experiment results indicates that the template matching method based on texture feature extraction demonstrates higher efficiency,with matching accuracy consistently above 96%.
关键词
图像特征/模板匹配技术/Hu不变矩特征/统计特征/纹理特征
Key words
image features/template matching techniques/Hu invariant moment feature/statistical feature/texture feature
维普
万方数据