An image contour recognition algorithm based on improved Hu moment invariants and SVM

Aiming at the problem that the existing Hu moment invariant operator is sensitive to scale changes and has low accuracy when dealing with intra-class changes with geometric transformations (rotation and large scale changes), a image contour recognition algorithm based on improved Hu moment invariants and SVM is proposed. Firstly, the Canny operator is used to obtain the contour of the image to be recognized, and then a scale-invariant Hu moment invariant algorithm is designed. The support vector machine classifier is trained and tested based on the Hu moment invariant of the shape contour. Experimental results show that this method has good robustness and high shape classification accuracy when dealing with geometric transformations such as translation and scaling.