영상개선을 위한 가중 메디안 혼성필터에 관한 연구
- Alternative Title
- A study on the weighted median hybrid filter preserving image details
- Abstract
- The weighted median hybrid filters are proposed. These filters take weighting factors in calculatingly the output of subfilters of median hybrid filters. The properties of weighted median hybrid filters are analysed and compared to the conventional median filter. The results indicate that these filters preserve details of edge better than and are computationally much more efficient than conventional median filter.
Performances of some image restoration filters(i.e. weighted median hybrid filter, median hybrid filter, median filter and K-nearest neighbor averaging filter) are compared and evaluated on SNR gain and subject test. Test images used here are corrupted by Gaussian noise. Experimental results show that the weighed median hybrid filter among them are comparatively superior to others.
The weighted median hybrid filters are proposed. These filters take weighting factors in calculatingly the output of subfilters of median hybrid filters. The properties of weighted median hybrid filters are analysed and compared to the conventional median filter. The results indicate that these filters preserve details of edge better than and are computationally much more efficient than conventional median filter.
Performances of some image restoration filters(i.e. weighted median hybrid filter, median hybrid filter, median filter and K-nearest neighbor averaging filter) are compared and evaluated on SNR gain and subject test. Test images used here are corrupted by Gaussian noise. Experimental results show that the weighed median hybrid filter among them are comparatively superior to others.
- Author(s)
- 허수진
- Issued Date
- 1993
- Type
- Research Laboratory
- URI
- https://oak.ulsan.ac.kr/handle/2021.oak/5280
http://ulsan.dcollection.net/jsp/common/DcLoOrgPer.jsp?sItemId=000002024452
- Authorize & License
-
- Files in This Item:
-
Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.