Image Quasi-feature Point Extraction Algorithm

Image registration is a fundamental task in image processing used to match two or more images, for example, at different time, from different sensors or from different viewpoints. Registration is the foundation of many image processing and applications such as object recognition, change detection, 3-D modeling.The image registration method has very many kinds. Based on the image characteristic image registration is in the registration the most common method. Based on in characteristic image registration , characteristic main aim at characteristic point. Characteristic point extraction is the image registration essential step. In this paper the multianalysis has contrasted the Moravec, Plessey , susan operator Forstner operator,Trajkovic operator. This paper has made 8-domain improvement to the Trajkovic operator, from extraction efficiency, operator stability, Localization accuracy, Anti- noise, in the counting yield to the operator to carry on the analysis to compare to the operator with the test image to carry on the experimental analysis one by one.Aim at the operator research in the characteristic point in the foundation in view of based on the characteristic point image registration realization. In multianalysis image registration each essential step. This paper uses the entropy to divide the size equal thin grid to the image, extraction the characteristic point to each grid information content, Therefore in matching process, The information content compares the wide range the precision to be high, Has also guaranteed the basic precision request in the information content few regions. This article uses regarding the control point match thickly matches first fine matches again, Thus enhancement match efficiency, finally uses the afflne transformation realization image matching and so on. Finally has carried on the matching experiment. May see through the experimental result analysis, this method matching precision, the reliability are all high, Can carry on the effective registration to the image.