Abstract: Aiming at the problems of high computational complexity and low detection rate of current computergenerated image recognition algorithm, we proposed a computergenerated image recognition algorithm based on improved local binary pattern (LBP) and gradient feature. The improved algorithm was mainly based on the local texture feature of the image. Firstly, the feature vectors of the computergenerated images and the natural images were extracted, and then these features were classified by SVM classifier. The experimental results show that the proposed algorithm can effectively discriminate computergenerated images and real images.

Key words: image processing, local binary pattern, image gradient,  , image recognition algorithm, particle image velocimetry technology