Image compensating method and image compensating device in video mosaics

Abstract

The invention provides an image compensating method and an image compensating device in video mosaics, which are used for a writing mosaics method. Based on various paths of source videos, the following steps are respectively executed: storage space Yi, in which an i-th path source video is currently written, is recorded in real time, and storage space X, in which a source video is written the fastest is written, is recorded; when the recorded storage space X changes, if the storage space Yi recorded for the i-th path source video is determined to be spaced by Z storage spaces away from the currently-recorded storage space X, then an image which is written by the i-th path source video into the storage space Yi is copied and written into the spaced Z storage spaces, wherein Z is a preset integer and less than M, and M is the number of the storage spaces spaced between a read storage space and the storage space of the source video with fastest writing. The method and the device can effectively overcome image withdrawal in the writing mosaics method and improve the visual effect.

Description

technical field [0001] The invention relates to image processing technology, in particular to an image compensation method and device in video splicing. Background technique [0002] In application fields such as video surveillance and video conferencing, video signals from multiple sources are often synthesized into one output, so as to present a multi-screen effect for users on the same display terminal. For example, in a video surveillance system, video images from multiple cameras can be spliced ​​into multiple images and displayed on one monitor. [0003] Among the ways of implementing multi-picture splicing by using Field Programmable Gate Array (FPGA, Field Programmable Gate Array), the writing splicing method is widely used because it is easy to implement and is not affected by the video format. The implementation characteristics of write splicing are: control writing, sequential reading, specifically: divide the storage space into more than two sub-blocks, one sour...

AI Technical Summary

Problems solved by technology

like figure 2 As shown, in the process of sequentially writing the images of the four source videos into each storage space, at time t4, if the image of the first source video is delayed due to network delay, it may arrive at time t5, In order to keep writing consistent, the image p1_t4 arriving at time t5 is written into storage space 1, thus skipping the storage space with frame number 0, so that the sub-frame corresponding to the first source video in the storage space with frame number 0 The storage space has not been refreshed, and the historical image p1_t0 (shown in the shaded part in the figure) is still retained, which causes a rollback phenomenon, making the displayed video not smooth and affecting the visual effect

Images