Image matching is of great importance in many real-world applications. Although many traditional methods exist for image matching but they have several problems such as high computational cost problem and problem of scale invariance. For correct image matching extracting appropriate point pairs have great significance to establish the relationship for among the images to be compared. An automatic extracting method point pairs is proposed in this work. Firstly, it determines the common areas of two images and divides them into blocks uniformly. The corresponding blocks of the two images are marked with the same sequence number to establish the one-to-one regional matching relation. Then, for each block, k-means based nearest neighbor approach adopted to detect more relevant point pairs. Finally the regional matching strategy is performed to match the acquired points. The proposed method can extract control point pairs with high precision, and these point pairs conduct to improve the precision of image matching.