Abstract: A hand-distribution-features-based approach to hand gesture recognition is presented in this paper.Firstly,a segmentation scheme for images under complex background is proposed which is based on the proposed Brightness-Gauss-Model.Then,search-window is used to select the valid hand gesture and extract the hand distribution features,which are composed of the destiny distribution feature and the figures features.Finally,we integrated all the features to calculate similarity distance.In this paper,random sampling is introduced which can improve recognition rate.Even if there are the hand and the face in our Video Sequence Images,we can still recognize the hand gestures because of the using of search window.The experimental results show that if the ambient light is relatively stable,the algorithm proposed in this paper can recognize hand gestures in real-time with strong robustness,and it is invariant to rotation,scale and translation.For those gestures with large differences,the recognition rate is up to 98%.