Research on Computer Vision Target Detection Algorithm Based on Feature Recognition

Computer vision has formed many research branches because of its different application fields and uses. Due to the influence of various environmental factors in real life, the clarity, completeness and distinctiveness of such target features are lacking to a certain extent, which brings great challenges to the existing algorithms. In this paper, a computer vision target detection algorithm based on SIFT (Scale Invariant Feature Transform) feature recognition is proposed. Firstly, the algorithm preprocesses the target image and the input image, then detects the existence of some feature points, that is, generates feature vectors, and obtains candidate matching points through the matching of feature vectors. Finally, it judges whether the target is recognized or not according to the number of matching points. The research results show that the target detection algorithm proposed in this paper can effectively combine and filter matching through multiple steps, and the matching accuracy is significantly improved compared with the traditional algorithm. The research results show that the background subtraction method is used to extract the moving target, which ensures the integrity of the detected target and improves the real-time and accuracy of the moving target detection.