Combined Demosaicing and Adaptive Filtering on CFA Images

Author

Tajbakhsh, Touraj ; Grigat, Rolf-Rainer

Author_Institution

Hamburg Univ. of Technol.

fYear

0

fDate

0-0 0

Firstpage

1

Lastpage

6

Abstract

The gate-count of image signal processing (ISP) units in digital cameras is significantly dominated by the amount of line buffers utilized by the CFA (color filter array) image interpolation algorithm. Subsequent image filtering on RGB data requires further line buffers for each color channel. We present a local-adaptive Wiener filter for noise cancellation and image enhancement with incorporation of Chang´s variable number of gradients interpolation method deploying a 5times5 neighborhood for each pixel. Four line buffers in total are utilized for demosaicing and filtering. Our experimental results indicate a significant increase in SNR whereas edges are preserved and image contrast is enhanced

Keywords

Wiener filters; adaptive filters; cameras; gradient methods; image colour analysis; image enhancement; image segmentation; interference suppression; interpolation; CFA image; Chang´s variable number; ISP units; RGB data; color filter array; combined demosaicing; digital cameras; gradient interpolation method; image enhancement; image filtering; image signal processing; line buffer; local-adaptive Wiener filter; noise cancellation; Adaptive filters; Adaptive signal processing; Array signal processing; Colored noise; Digital cameras; Digital signal processing; Filtering; Interpolation; Signal processing algorithms; Wiener filter; combined demosaicing; denoise; filtering; image enhancement; noise reduction; sharpening;

fLanguage

English

Publisher

ieee

Conference_Titel

Consumer Electronics, 2006. ISCE '06. 2006 IEEE Tenth International Symposium on

Conference_Location

St. Petersburg

Print_ISBN

1-4244-0216-6

Type

conf

DOI

10.1109/ISCE.2006.1689422

Filename

1689422