A black level correction method for an image sensor

Abstract

The invention discloses a method for correcting the black level of an image sensor, comprising the following steps: S01: adopting a high-speed correction mode to sequentially correct the first frame of image data to the Nth frame of image data, S02: adopting a low-speed correction mode to sequentially correct the first frame of image data Correct the image data from the N+1 frame to the M frame image data, S03: Repeat steps S01-S02 until all the image data generated by the image generating module are corrected. The method for correcting the black level of an image sensor provided by the present invention can effectively correct the black level in the effective pixel area through the joint correction of the high-speed correction mode and the low-speed correction mode, and can make the color cast and flickering phenomenon in the image get better improve.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a black level correction method of an image sensor. Background technique [0002] In the process of image processing, the image collected by the CCD / CMOS image sensor is usually processed. However, when an image sensor is used to capture images, due to differences in the physical structure and process level of the four color channels of the sensor, and because of the influence of exposure time and filters, the black levels generated in the four color channels of the image sensor are inconsistent. , even if the image sensor is completely blacked out and not sensitive to light, the four color channels will have differences due to the influence of the ADC output. At the same time, factors such as the operating temperature, exposure time, and operating voltage of the image sensor will also affect the black level values ​​of the four channels, resulting in unfavorable conditions such ...

AI Technical Summary

Problems solved by technology

[0003] The traditional method is to count the average value of the black level pixels without gain in the black level area of ​​the image (that is, the non-sensitive pixel area) in the ASP module regardless of the channel, and then feed back the average value to the current frame image for black Level correction. The disadvantage of this method is that when the temperature or exposure changes, the black level will change accordingly. This change is not completely linear, and the higher the temperature, the higher the image signal and the black level. The more obvious the overlap, it is manifested as obvious dark noise. In this case, the average value of the pixels in the black level area of ​​the current frame cannot really represent the black level in the current environment, and it is easy to generate large errors. Therefore, the image is corrected in the black level It is easy to produce flickering phenomenon in the process of

[0004] In addition, there are methods to set the adjustment step and calculate the correction target threshold of the current frame image and set it as the correction reference value, and then use negative feedback to make the feedback value gradually adjust to the correction according to the set adjustment step in each frame. Reference value, and feed back to the current frame image at the same time for black level correction, this method to a certain extent improves the flickering phenomenon of the image screen caused by the sudden change of the correction value, but because only reference to the correction reference value, when this value is inaccurate , it is easy to cause color cast in the image

These methods first ignore the differences between channels when calculating the average value of the black level area, and all of them are processed in the ASP module, without considering the influence of the gain of the image sensor itself, resulting in the black level of a certain channel. If the leveling correction is not in place, it is easy to cause bad situations such as color cast, flickering or over-correction of the image

Images