In order to effectively remove high dense impulse noise in images,a novel two-stage impulse noise filtering algorithm based on improved Pulse Coupled Neural Networks(PCNN)noise detection is put forward.Firstly,the noisy points and signal points are distinguished and located using the synchronous pulse burst property of PCNN,then the noisy points are smoothed by the first adaptive filter based on their local adjacent information of the noisy points,finally,the assistant filtering using the second multi-direc...