Application of Adaptive Neighborhood Denoising Algorithm in Image Enhancement
Lei Lei, Liang Wang, Qiannan Xue, Huanlang Wang, Yiqing Shi, Xiaohui Dai · 2023
As the basis of image processing, image denoising has always been a research hotspot in the field of computer vision. The purpose of image denoising is to suppress noise from the distorted image and restore the image as much as possible, so that the restored surround image is closer to the original image in quality. Aiming at the Neigh Shrink method used in wavelet image denoising, this paper proposes an effective image denoising algorithm to protect the image edge. Mainly improving the fixed neighborhood range in the Neighbor Shrink method, adaptively segmenting the image into different neighborhoods for denoising based on its own properties; And further combining the correlation within the wavelet layer, fixed windows were added to each irregular neighborhood, and coefficients with geometric distances closer and within the same irregular neighborhood were selected to improve the Neigh Shrink method. We studied image enhancement based on wavelet transform, first analyzed the basic methods of image enhancement, and then put image enhancement into the wavelet domain for research. We proposed a new adaptive image enhancement algorithm using wavelet transform, and verified its feasibility and superiority in the environment.