Abstract: In the process of image denoising, using the BP algorithm to establish the model which has the initial weight threshold is random, and the model is easy to fall into the local minimum.This paper presents an idea which is using bat algorithm to optimize BP algorithm model weight and threshold parameter, the image denoising model is based on the bat BP neural network algorithm. By comparing with Wiener filtering, BP model and particle swarm BP model image denoising effect, the bats BP neural network denoising model has better structural similarity and peak signal to noise ratio after denoising the image. It is proved that the de-noising model based on the bat neural network algorithm has better denoising effect.