Method and device for demosaicking

Abstract

The invention relates to a method and a device for demosaicking. The method comprises the following steps of: acquiring a Bayer-format image; carrying out gradient extraction on a pixel array which takes a target pixel point as the center; carrying out weight extraction on the basis of a gradient extraction result to obtain a weight coefficient of the pixel array; on the basis of the weight coefficient, simultaneously carrying out color reconstruction on other two colors missed by the target pixel point; and outputting image information. The device comprises an image acquisition unit, a gradient extraction unit, a weight extraction unit, a color reconstruction unit and an image output unit, wherein the image acquisition unit is used for acquiring the Bayer-format image; the gradient extraction unit is used for carrying out gradient extraction on the pixel array which takes the target pixel point as the center; the weight extraction unit is used for carrying out weight extraction according to the gradient extraction result to obtain the weight coefficient of the pixel array; the color reconstruction unit comprises two color reconstructors connected with the weight extraction unit and is used for simultaneously carrying out color reconstruction on other two colors missed by the target pixel point according to the weight coefficient; and the image output unit is used for outputting the image information obtained after color reconstruction. According to the invention, the image processing efficiency can be improved.

Description

technical field [0001] The present invention relates to the field of image processing, and in particular to a method and device for demosaicing. Background technique [0002] Image sensors such as Complementary Metal Oxide Silicon (CMOS) or Charge-Coupled Device (CCD) are used in digital cameras, notebooks, tablet computers, mobile phones and other electronic devices that can capture digital images. To capture images, the image sensor is used to convert optical images into digital signals. [0003] Each pixel in a digital image usually contains three components of red R, green G and blue B. Each photosensitive unit of a single CMOS or CCD image sensor can capture a light with a specific wavelength, such as R, G, One of B. In order to obtain a complete image including R, G, and B components, electronic equipment needs to use three CMOS or CCD image sensors to capture R, G, and B components at each pixel. This increases the cost of the electronics and also increases the siz...

AI Technical Summary

Problems solved by technology

[0019] The existing demosaic method performs color reconstruction on green, red (or blue) sequentially when performing color reconstruction, which is serial processing and requires more processing steps and processing time

Images