Image processing method, device and server

Abstract

The embodiment of the invention discloses an image processing method, device and server, which are used to improve the coincidence degree of a target frame and a foreground target. In the embodiment of the present invention, first obtain the input frame and the background frame, then determine the row or column to be cut in the target frame, calculate the gradient difference matrix of the input frame and the background frame in the row or column to be cut, and calculate the gradient difference matrix The average value of the gradient difference, judge whether the average value is less than the gradient comparison value, if it is less than, cut off the position to be cut from the target frame. In the embodiment of the present invention, the similarity between the input frame and the background frame is used as the basis for cutting the target frame, and the average value of the gradient difference value of the row or column in the target frame of the input frame and the background frame is respectively calculated as a comparison feature, and compared with the gradient comparison value , so as to judge whether the current row or column is a real foreground target, if not, cut off the current row or column from the target frame. In this way, the target frame can be effectively corrected, and the coincidence between the target frame and the foreground target can be improved.

Description

Image processing method, device and server technical field [0001] The present application relates to the computer field, in particular to an image processing method, device and server. Background technique [0002] In the application of image video sequences, people are usually only interested in some parts of them, such as video surveillance. In the system, we care about the foreground targets with more significant motion, but do not care about the relatively static background targets, so We need to distinguish the foreground from the background well, and get a target frame that can accurately frame the foreground target. In practical application scenarios, due to the influence of illumination and other reasons, the foreground target area obtained by the target detection algorithm It is often larger than the actual foreground target, or generates some false targets, which affect subsequent application scenarios. For example, in the target tracking stage Segment, wh...

AI Technical Summary

Problems solved by technology

In the actual cutting process, due to the influence of light, it is difficult to obtain a more accurate target frame.

Images