Analysis On the Cause of Moiré Effect
Capturing reality and accurate colors have always been one of the biggest challenges in digital imaging. Because black and white sensors can only measure the intensity of light. So we designed a method to measure the intensity of light of a particular color on each pixel. This method involves placing a small Bayer mode small filter of alternating colors on each pixel.The Bayer pattern consists of a red and green alternating filter of even line pixels, a green and blue alternating filter of odd line pixels. Because the human eye is most sensitive to green light, half of the sensors (pixels) are sensitive to the wavelength in this range. It is worth noting that there are other modes of filters, such as sparse color filter arrays, but far less popular than Bayer filters. Extracting a complete color image is a challenge which from a mosaic-mode sensor consisting of 50% green light, 25% red light, and 25% blue light. The concept is known as demosaicing, which involves estimating the two missing color values of the pixel using the color values of the surrounding two pixels. The easiest way is a linear interpolation. In the case of a green pixel, the average of the adjacent red pixels can be assigned to the red component of the green pixel, as is the case for adjacent blue pixels. A more complicated method is to take the average of the red and blue components in a 5×5 square around the green pixel to get a more accurate color.
In general, these methods are better, but some images will have image artifacts that are not in the original scene. Some of these artifacts are visible at the original resolution of the image, while others need to be enlarged to become apparent. False coloring is a type of artifact that is due to a difference error near the edge of the object, or a color that is not in the original scene. Another similar artifact is the Moiré effect, which often appears on high-frequency patterns, such as striped shirts. It is a new pattern of twisting due to the interpolation algorithm used. (see Figure 4).The third type is called zippering. Zipper artifacts due to the mean value of the de-mosaic algorithm around the edge of the object. Once the image is enlarged, the effect of this stair or zipper pattern along the edge becomes evident (see Figure 5).