New locally adaptive image interpolation algorithm based on edge preserving

In order to reduce the influence of noise and preserve edge information, a new locally adaptive image interpolation algorithm based on edge preserving is proposed. First, integer wavelet transform is applied to the partially overlapped sub-images. Then, domain BayesShrink threshold denoising and median filtering are used in the high frequency band to remove Gaussian noise and impulse noise, and brightness preserving histogram equalization is used to enhance the contrast of the image in low frequency band. Finally, the sub-images are segmented into flat areas and edge areas using self-adaptive threshold Canny edge detection, and then bilinear interpolation and nonlinear interpolation are carried out. Experiment results show that, the obtained interpolated image using MBW interpolation algorithm has good subjective image quality, and the average PSNR is improved by 2.23 to 2.96 dB compared with those of traditional interpolation methods.