1. 변형된 마스크와 유사 화소 간 그룹화를 통한 S&P 잡음 영상 복원.
- Author
-
정영수 and 김남호
- Subjects
MAGNETIC resonance imaging ,IMAGE processing ,COMPUTED tomography ,INTERNET of things ,FILTER paper ,IMAGE denoising ,NOISE ,HOUGH transforms - Abstract
With the advancement of IoT(Internet of Things), image processing is being studied in various fields. Salt and pepper(S&P) noise in image acquisition devices, such as CT(Computed Tomography), MRI(Magnetic Resonance Imaging), X-ray, and scanning equipment, may develop due to environmental and physical factors, thereby distorting the original image. Numerous studies have been conducted to remove S&P noise, but existing filters have shown significantly insufficient restoration of high-density noise, low-light, and patterned images. Therefore, this paper proposes a filter high-density and low-light noise environments and preserve the patterns properties appearing in the images. In the proposed algorithm, we design local masks transformed for high-density noise and low-light images and defined the patterns represented in the images by determining and grouping similar pixels. Different filters are applied according to the pattern of similar pixel group to preserve the geometry based on its position in the input image. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF