The work aims to improve the matching accuracy and robustness of image matching algorithm. The image matching algorithm based on region adaptive model coupled vector constraint rule was designed. First, the saliency analysis method using context information was introduced to extract the salient regions and non-significant regions of the image. According to the saliency of the region, a region adaptive model was constructed to dynamically adjust the gray threshold in the FAST algorithm and extract the feature points in the image. Then, the points in the neighborhood of the feature point were divided into long and short point sets by the Euclidean metric. The feature direction was generated by the long point set, and the feature vectors were generated by the short point set to obtain the descriptors of the feature points. Finally, the Hamming distance between feature points was measured to achieve the matching of feature points. The vector constraint rules were established by means of the vectors composed of matching feature points to optimize the matching feature points, so as to complete the image matching. The experimental results showed that, compared with the current image matching technology, the proposed algorithm had higher robustness and matching accuracy. When the target rotation angle reached 100°, its matching accuracy could still reach 88.95%. With good adaptation and better matching accuracy in the case of geometric transformation, the proposed algorithm has good reference value in the fields of image processing, information security and so on.