Aiming at the problem of automatic traffic signs recognition, the method of circle signs image recognition based on invariant moments and support vector machine (SVM) is studied in this paper. Firstly, according to the color and shape information of traffic sign, the original image is processed by color segmentation, morphological denoising and shape detection. Then, Hu and Zernike invariant moments of the images are extracted to establish the corresponding feature data set, and the data set is input into SVM and the grid search technique is used to optimize the parameters of SVM. Finally, the traffic signs are recognized in the trained SVM classifier. The experimental results show that compared with other existing traffic signs recognition algorithms, the high order Zernike moments and the optimized SVM have better recognition results.