Hybrid compression of CFA images using edge based segmentation

Single sensor cameras use color filter arrays(CFA) to capture the images. The sensors in a CFA are arranged to capture any one of the Red, Green or Blue components of a pixel in an interleaved way. The bit rates attained while compressing the captured CFA images are high for all the standard compression methods. Different strategy has to be adopted for compressing CFA images due to the unevenness caused by interleaving of components. This paper proposes a method by which the CFA images are smoothed by 5/3 wavelet transform and compressed by conserving the central subject. The background of the image is kept with certain level of loss. This method reduces the bit rate comparing to the standard methods.