Abstract: Combining with the advantages of curvelet transform and TV model,proposes a denoising method based on curvelet transform and TV model,which detects the edges of the image with the high-frequency coefficients of curvelet transform,introduces the edge detecting function to guide the TV model diffusion,the new model can not only remove noise, but protect the edges of the image.use peak signal-to-noise ratio to evaluate the denoising results,the experimental results show that the denoising effect of the new model is significantly better than that of the curvelet transform threshold method and TV model.