Edge detection from Bayer color filter array image

Shen-Castan Based Edge Detection Methods for Bayer CFA Images

“…In the works of Aberkane et al [4] and Magnier et al [5], Deriche-based algorithms were proposed (Eqs. ( 5) and ( 6)) directly on the CFA image to extract contours.…”

“…However, it is possible to obtain a seamless image in exchange for the image resolution by down-sampling the pixels of the same color channel from a CFA image and then reconstructing the image without empty values. To solve this, Aberkane et al [4] introduced a spatial distance parameter (d) directly into the Deriche filter implementation (Eqs. ( 5) and ( 6)), so that the filters can be applied on CFA images the same way as on a full image (image without empty values).…”

“…Consequently, for simplicity and convenience, features may be extracted using grayscale vision sensors indicating only the brightness of a pixel. In response to the demand of feature extraction efficiency directly on CFA images, Aberkane et al [4] and Magnier et al [5] have proposed edge detection algorithms that operate based on Deriche edge detector and bilinear interpolations [6].…”

A Recursive Edge Detector For Color Filter Array Image

“…The recursive Deriche filter Deriche [10] created an optimal edge detection filter according to Canny's criteria [11]: unicity, good detection and good localization. This filter can be implemented recursively allowing reducing computation time [8]. Deriche used a gradient approach based on two separable filters for the computation of partial derivative images: (i) a smoothing filter s α (t) approximating a Gaussian function, (ii) a derivative filter d α (t) which is the derivative of s α (t):…”

“…Aberkane et al [7] showed that directly edge detection in those images is possible with a very close or even better precision than segmentation after demosaicking. They introduced two approaches to use the raw data of CFA images for edge detection: (i) Deriche-based Luminance (DL) technique estimates partial derivative images of luminance thanks to Deriche recursive filters [8], so that a gradient image is computed, (ii) Deriche-based Color (DC) derivative method also uses Deriche filters to compute partial derivatives of red, green and blue channels so that a structure tensor is applied [9]. Despite the best results offered by DL, methods computing luminance derivatives from Bayer CFA were not studied deeply enough.…”