IBM3D: Integer BM3D For Efficient Image Denoising

Image denoising is the basis of image processing procedure and can improve the quality of the noisy image,better image denoising performance leading to better results for the following image processing steps.The block-matching collaborative filtering(BM3D)denoiser has been considered as a strong performer in image denoising,but it has high computational cost and requires a large amount storage space in block-matching and 3D transforms,which limits its practical applications,particularly in embedded video processing systems.In this paper,we propose an integer BM3D(IBM3D)that involves only integer operations.To integerize 3D transforms and Kaiser window,balance between approximation accuracy and denoising performance are carefully investigated.We propose an integer Wiener filter and investigate its performance over the original empirical Wiener filter with both analytical analysis and experimental verifications.We use the butterfly computation to optimize the DCT transform and Hadamard transform.The experiment results show that the proposed IBM3 D provides comparable denoising performance to the original BM3 D for a wide range of noise levels,and generates even better results for high noise levels.The proposed IBM3 D requires less computation than the original BM3 D,and can be deployed into embedded systems without or with limited floating-point computation resources,and ported to chips with smaller circuit areas and less power consumption.