Method and device for white balance calibration and method and device for obtaining calibration parameters

Abstract

The invention relates to a white balance calibration method and device and a calibration parameter acquisition method and device thereof. The method for obtaining white balance calibration parameters includes the steps of: obtaining the eyeball area in the image to be calibrated; obtaining the RGB ratio of the eyeball area according to the eyeball area; determining whether the eyeball area is gray according to the RGB ratio; if so, determining the RGB ratio as a white balance calibration parameter . The white balance calibration parameter acquisition device includes: an eyeball area acquisition module, used to acquire the eyeball area in the image to be calibrated; an RGB ratio acquisition module, used to obtain the RGB ratio of the eyeball area according to the eyeball area; a gray detection module, used to obtain the RGB ratio of the eyeball area according to the RGB The ratio determines whether the eyeball area is gray; the white balance calibration parameter determination module is configured to determine the RGB ratio as the white balance calibration parameter when the eyeball area is gray. The invention can accurately restore the white balance without being affected by ambient light and background.

Description

Technical field [0001] The invention relates to the technical field of white balance, in particular to a white balance calibration parameter obtaining method, a white balance calibration parameter obtaining device, a white balance calibration method and a white balance calibration device. Background technique [0002] White balance is a very important concept in the field of TV photography, through which a series of problems of color reproduction and tone processing can be solved. The most basic principle of white balance is to restore white things to white. However, after the sensor outputs the image, the ISP (Image Signal Processor) does not know which area itself is white, and can only count the R / G (red / green), B / G of the entire image (Blue / green), and then use the algorithm to estimate which ones need to be restored to R:G:B=1:1:1. In this way, once there is a mixed light scene or a single-color object, the white balance will have defects such as inaccurate restoration. S...

AI Technical Summary

Problems solved by technology

In this way, once there is a scene with mixed light or an object with a single color, the white balance will have defects such as inaccurate restoration.

Images