Reference frame updating method and system in grayscale projection image stabilization

Abstract

The invention discloses a reference frame updating method and system in grayscale projection image stabilization. The reference frame updating method in the grayscale projection image stabilization comprises the following steps: obtaining a current frame and a reference frame of the current frame, and obtaining a grayscale projection histogram of the current frame and the reference frame of the current frame, wherein the grayscale projection histogram of the current frame is represented by S1, and the grayscale projection histogram of the reference frame of the current frame is represented byS2; matching the S1 and the S2 through moving the S1 to obtain the translation quantity of the S1; obtaining the difference image of the current frame and the reference frame; calculating the pixel point average value of the difference image according to the difference image; obtaining a threshold value of the pixel point average value of the difference image; and judging whether pixel point average value of the difference image is larger than the threshold value of the pixel point average value of the difference image or not, so as to judge whether the current frame is used as the reference frame for the next frame image or not. According to the reference frame updating method and system in the grayscale projection image stabilization, the reference frame is updated according to the actual situation; therefore, the disadvantages of failure of method, frequent updating of the reference frame, waste of computing resources and decline of computing speed caused by not in time updating ofthe reference frame are avoided.

Description

technical field [0001] The invention relates to the field of image stabilization, in particular to a method and system for updating reference frames in grayscale projection image stabilization. Background technique [0002] The traditional image stabilization algorithm based on grayscale projection needs to crop the original image, sacrificing part of the image information, and reducing the visible range of the camera. In addition, the reference frame in the traditional algorithm is fixed, and the difference between the current frame and the reference frame is too large. The time cannot be matched, the algorithm is invalid, and the scenery is constantly changing during the driving process. The traditional image stabilization algorithm does not update the reference frame. When the scenery is greatly different from the initial moment, the current frame cannot be stabilized in real time. Road vehicles, with large road surface roughness and severe vehicle bumps, require a large ...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a method and system for updating reference frames in gray-scale projection image stabilization, to solve the problem of failure of the method in the prior art when the update of the reference frames is not timely, and waste of computing resources when the update of the reference frames is too frequent. The problem that the calculation speed drops, the matching error, and the image stabilization effect is greatly reduced

Images