Image registration ghost elimination method based on optical flow mapping repetitive region detection

Abstract

The invention discloses an image registration ghost elimination method based on optical flow mapping repetitive region detection. The method comprises the following steps: pre-registering two images by using an optical flow between a reference frame image and a frame image to be registered; recording the corresponding position of each pixel in the image to be registered in the pre-registration result; taking a corresponding position of a pixel which is used for multiple times in the image to be registered in the pre-registration result as a suspected ghost position; and removing the pixel mostsimilar to the reference frame image from the plurality of candidate ghost image pixels, determining the rest pixels as ghosts, and finally, replacing the pre-registration result by using the corresponding pixel of the reference frame image at the detected ghost position, thereby obtaining a ghost-free image registration result. The method is high in precision and high in efficiency, can effectively remove the ghosting without increasing the time complexity of the registration algorithm, and is a practical method and means capable of realizing image registration and ghosting removal in an actual camera imaging system.

Description

technical field [0001] The invention belongs to the technical field of image registration in digital image processing technology, and in particular relates to an image registration ghost removal method based on optical flow mapping repeated region detection. Background technique [0002] Image multi-frame fusion technology is widely used in smartphone image processing tasks such as HDR, multi-frame noise reduction, and image super-resolution. Image registration is an essential module of multi-frame fusion algorithm. The traditional global registration method calculates a global transformation matrix based on feature point matching to transform the image. However, it is limited by the fact that there may be different transformation relationships between parts of the image and there may be moving objects in the image. The output of the traditional global registration algorithm The results are flawed. [0003] Dense optical flow outputs the displacement of each pixel in two f...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problems in the background technology, the present invention proposes an image registration ghost removal method based on optical flow mapping repeated region detection, which can effectively solve the current difference-based ghost removal algorithm commonly used in optical flow registration. Technical defects of low efficiency and poor robustness

Images