In order to improve the overall performance of embedding algorithm,th is paper proposes a reversible data hiding scheme based on edge detection and pixel class ification for gray images.In accordance with the priority principle of embedding capacity,the optim al threshold is chosen to extract the image edge,and according to the edge informat ion and the specified image quality control factors,the pixels are divided into smooth,weak edge and stron g edge pixels.For smooth pixels,by increasing the embedding strength to improve embedding capacit y,the prediction error histogram shifts by 2;For the weak edge pixels,the maximum modification i s 1,thus ensuring the quality of the stego-image;Strong edge pixel value remains unchan ged,further improving the stego-image quality.Experimental results show that the proposed a lgorithm can achieve higher embedding capacity and imperceptible distortion,and the overall performance of the algorithm is superior to that of other similar algorithms.