Image flicker removing method

Abstract

The invention discloses an image flicker removing method which is used for removing flicker of other frame images except a first frame image in a video image. The method comprises the following steps:substituting a previous frame flicker generation before image of a current frame and a current frame flicker generation after image into a linear model containing parameters, then obtaining the parameter values; substituting the parameter values into the linear model containing the parameters, then obtaining a model for removing the flicker; substituting the current frame flicker generation afterimage into the model with flicker removal, and obtaining a flicker generation before image of the current frame. The invention not only can effectively remove image flicker, but also can get rid of the dependence on experience of operators and has higher automation degree.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method for removing image flicker. Background technique [0002] Flicker refers to the random change of the video image in space and time that does not reflect the real brightness of the scene. This random change will not only give people a visual effect of one light and one dark, make people easy to fatigue, but also affect the video quality. Image compression. Therefore, it is necessary to remove the flicker appearing in the video image. Since there are many reasons for flickering and there are many manifestations of flickering, how to remove flickering becomes a very complicated problem. [0003] At present, the commonly used flicker removal method is realized by establishing a flicker model and estimating the parameters of the established flicker model, and the existing flicker model has two types: a nonlinear model and a linear model. The following two models ar...

AI Technical Summary

Problems solved by technology

This method can remove flicker to a certain extent, but there are also shortcomings: first, the histogram does not contain position information, so the effect of using this method to remove flicker is not obvious, especially when the image contains motion areas , the acquisition of the histogram is even more inaccurate, therefore, the flicker removal effect of the entire image will be greatly reduced; secondly, if there are large spots in the image, the impact on the histogram will be very large, so the estimation of the histogram The accuracy will also be greatly reduced, which will greatly reduce the flicker removal effect

However, the equation has I n (x, y), I' n (x, y), a(x, y) and b(x, y) 4 quantities, and I' n (x, y), a(x, y) and b(x, y) are all unknown quantities, so the equation is an ill-conditioned equation that cannot be solved at all, which restricts the use of linear models to remove the flicker generated by the image

Images