Moiré Reduction Using Inflection Point and Color Variation in Digital Camera of No Optical Low Pass Filter

Dae-Chul KIM  Wang-Jun KYUNG  Ho-Gun HA  Yeong-Ho HA  

IEICE TRANSACTIONS on Information and Systems   Vol.E98-D   No.12   pp.2290-2298
Publication Date: 2015/12/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.2015EDP7201
Type of Manuscript: PAPER
Category: Image Processing and Video Processing
optical low pass filter,  moiré reduction,  color smearing,  image quality,  

Full Text: PDF(1.4MB)>>
Buy this Article

The role of an optical low-pass filter (OLPF) in a digital still camera is to remove the high spatial frequencies that cause aliasing, thereby enhancing the image quality. However, this also causes some loss of detail. Yet, when an image is captured without the OLPF, moiré generally appears in the high spatial frequency region of the image. Accordingly, this paper presents a moiré reduction method that allows omission of the OLPF. Since most digital still cameras use a CCD or a CMOS with a Bayer pattern, moiré patterns and color artifacts are simultaneously induced by aliasing at high spatial frequencies. Therefore, in this study, moiré reduction is performed in both the luminance channel to remove the moiré patterns and the color channel to reduce color smearing. To detect the moiré patterns, the spatial frequency response (SFR) of the camera is first analyzed. The moiré regions are identified using patterns related to the SFR of the camera and then analyzed in the frequency domain. The moiré patterns are reduced by removing their frequency components, represented by the inflection point between the high-frequency and DC components in the moiré region. To reduce the color smearing, color changing regions are detected using the color variation ratios for the RGB channels and then corrected by multiplying with the average surrounding colors. Experiments confirm that the proposed method is able to reduce the moiré in both the luminance and color channels, while also preserving the detail.