Method and device for real-time fusion of multiple video images

Abstract

The embodiment of the invention provides a multi-channel video image real-time fusion method and device, and the method comprises the steps: storing a target pixel which is the pixel with the minimumcoordinate value in all paths of currently processed pixels, and enabling the coordinate value of the pixel to be obtained through the calculation according to the coordinate value of the pixel and apreset algorithm model; And if it is judged that the storage space corresponding to the coordinate value of the target pixel is full, fusing the pixels stored in the storage space. The embodiment of the invention provides a multi-channel video image real-time fusion method and device. multiple paths of high-definition video data can be cached in real time; According to the invention, the multi-channel video data is received and fused at the pixel level, and the fused data is sent out in real time, so that the problem that the processing speed and the real-time performance of the traditional embedded equipment are limited is effectively solved, and the video stream data is successfully cached and fused and transmitted on the basis of not limiting the transmission speed of the video stream data.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of image processing, and in particular to a method and device for real-time fusion of multi-channel video images. Background technique [0002] With the development of big data and artificial intelligence algorithms, image recognition has been applied in more and more fields. Image recognition is the use of computers or embedded devices to fuse, analyze and understand images collected by sensors to recognize various patterns. The goal and technology of the object. When the embedded device recognizes and fuses the target, it needs to transmit and fuse a large amount of image data at a high speed. How to improve the efficiency of the embedded device to transmit and fuse the massive data collected by the sensor is to ensure that the embedded device is The key factors of high speed and real-time image fusion. [0003] The common method currently used is to store multi-channel high-definit...

AI Technical Summary

Problems solved by technology

[0003] The common method currently used is to store multi-channel high-definition real-time images in the buffer or memory in the format of pictures, and then read the data in the buffer for fusion after storing a full picture. When there are many data sources , such a storage and transmission mechanism seriously reduces the fusion speed of the embedded system, especially the characteristics of FPGA parallel fusion, this storage mechanism will make the parallel processor idle for a large amount of time, waiting for all the data to be stored in an image After the integration, this severely limits the fusion speed and real-time performance of embedded devices

Images