关键词: 脉冲耦合神经网络, 椒盐噪声, 图像去噪, 自适应滤波

Abstract: A new adaptive de-noising method is proposed based on improved Pulse Coupled Neural Network（PCNN）.Aimed at de-noising the image with serious salt & pepper noise effectively，preserving more image details，this method introduces a kind of detection mechanism of locating noised pixels based on the firing time map of PCNN，and only filters these pixels using an analogous median filter.Furthermore，it automatically selects the optimal filtering times and size of filtering window based on the estimated noise intensity to enhance the adaptability and de-noising ability of the system.Experiment results prove that the method based on the adaptive PCNN system can remove noise and preserve the details of images more effectively and completely than the conventional methods，and the strong adaptability is the main feature of the system.

Key words: pulse coupled neural network, salt &, pepper noise, image de-noising, adaptive filtering