| Abstrakt: |
As a medium for conveying visual information, video is applied in various fields. However, due to an environmental factor or external interference, various noises may appear in images, and salt and pepper noise is typical. In this paper, a homography based salt and pepper denoising method using bilinear interpolation is suggested to effectively remove such noise. Bilinear interpolation is used to estimate the data corrupted by noise and the values between the two axes are interpolated with the selected observing point by dividing a mask into four zones. In addition, a homography technique is used as an auxiliary means to get accurate estimates, and shows excellent visual and numerical performance compared to existing methods, and is expected to contribute to the improvement of image quality as a new noise removal method with strength against changes in the density of noise. [ABSTRACT FROM AUTHOR] |
