Investigation on colour sensor denoising
Bilateral filtering is a popular denoising technique that smoothes away noise while preserving edges by nonlinear averaging of similar adjacent pixel values. The original filter does not remove salt and pepper noise effectively and textures of the image are not well preserved. Thus, we propose a mod...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2011
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/44272 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Bilateral filtering is a popular denoising technique that smoothes away noise while preserving edges by nonlinear averaging of similar adjacent pixel values. The original filter does not remove salt and pepper noise effectively and textures of the image are not well preserved. Thus, we propose a modified form of bilateral filter that separately operates on high frequency and low frequency sub-band to extend the classical bilateral filtering for image noise removal. A new low frequency noise suppressor is incorporated into the modified bilateral filter framework. The proposed filter is shown to perform better than the traditional bilateral filter. The filter is implemented in Matlab into a graphical user interface with noise level estimation function and image blending function to target different noise level in the digital image. |
|---|