Abstract
Hot pixels are defective pixels in image sensors produced in the manufacturing process. The amount of these hot pixels is likely to be higher for higher resolution digital cameras, and this may increase with usage. These defective pixels are impulsive in nature and will cause severe colour artifact problems in the colour filter array (CFA) interpolation process for demosaicking of a full colour image. In our proposed method, novel modified bilateral filters are developed to exclude these defective pixels from interfering with the demosaicking process. Once the missing colour pixels are populated, a median based detail-preserving filter is then applied to remove hot pixels from the original sensor data. From our experimental results, it has been shown that our proposed method is able to exclude hot pixels within the demosaicking process in order to minimise colour artifacts. These results have visually and quantitatively confirmed that the proposed method outperforms other existing techniques.
| Original language | English |
|---|---|
| Pages | 178-183 |
| Number of pages | 6 |
| DOIs | |
| Publication status | Published - 1 Dec 2013 |
| Event | 28th International Conference on Image and Vision Conference on Image and Vision - Duration: 27 Nov 2013 → … |
Conference
| Conference | 28th International Conference on Image and Vision Conference on Image and Vision |
|---|---|
| Period | 27/11/13 → … |
UN SDGs
This output contributes to the following UN Sustainable Development Goals (SDGs)
-
SDG 9 Industry, Innovation, and Infrastructure
Fingerprint
Dive into the research topics of 'Hot Pixel Removal within Colour Filter Array Demosaicking using Modified Bilateral Filters'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver