Abstract:
A method for filtering the noise of a sequence of digital images (Img 1 ,..., Img n , Img n+1 ,...) in CFA format comprising the following phases:
processing a first CFA image (Img n-1 ) of the sequence to obtain a corresponding improved CFA image (f_Img n-1 ) with reduced noise; processing at least one pixel (p n (x,y)) of a second CFA image (Img n ) of the sequence that temporally follows said first image (Img n-1 ), said phase of processing at least one pixel providing a corresponding filtered pixel (f_p n (x,y)) and including the following operations:
selecting (SW_ sel) a first set of pixels (SW n (x,y)) comprising said at least one pixel and a plurality of pixels of the second CFA image spatially adjacent to it; selecting (TW_sel) a second set of pixels (TW n (x,y)) comprising pixels of the improved CFA image homologous with the pixels of said first set; carrying out a digital filtering of a first type (ST_fliter) using pixels forming part of said first and second set to generate the filtered pixel (f_p n (x,y)).
Abstract:
A method of filtering and an image filter (10) is disclosed. The filter is provided for a digital camera including image sensors sensitive to light, a color filter placed over sensitive elements of the sensors and patterned according to a Bayer mosaic pattern layout and an interpolation algorithm joining together the digital information provided by differently colored adjacent pixels in said Bayer pattern. The filter (10) is adaptive and includes a noise level computation block (26) for operating directly on a said Bayer pattern data set of for each color channel thus removing noise while simultaneously preserving picture detail.