Image processing device, image processing method, terminal and server

Abstract

The invention discloses an image processing device, an image processing method, a terminal and a server. The image processing device comprises a first cache module, a second cache module and an imageprocessor. The first cache module is used for receiving (N-2)th frame of image rendered by an image processing module through a second thread in the process of uploading Nth frame of image to the image processing module through a first thread. The second cache module is used for receiving and caching the rendered (N-1)th frame of image through a third thread and sending the (N-2)th frame of imageto the first cache module through the second thread in the process of receiving and caching the Nth frame of image uploaded by the first cache module through the first thread. The image processor is used for reading the (N-1)th frame of image from the second cache module through the third thread, rendering the (N-1)th frame of image, and writing the (N-1)th frame of image into the second cache module. According to the technical scheme, uploading of the Nth frame of image, rendering of the (N-1)th frame of image and loading of the (N-2)th frame of image can be carried out simultaneously so thatwaiting time for uploading and loading of the image can be saved and image processing speed can be increased.

Description

technical field [0001] The present invention relates to the field of image technology, in particular to an image processing method, device, terminal and server. Background technique [0002] At present, with the increasing demand of users for real-time video processing, real-time image processing has always been the bottleneck in the process of real-time transcoding. Therefore, how to more effectively improve the speed of image processing is a major problem in realizing real-time video transcoding. In related technologies, the image is usually processed based on GPU (Graphics Processing Unit, graphics processor) through opengl (OpenGraphics Library, which refers to a professional graphics program interface that defines a cross-programming language, cross-platform programming interface specification), However, due to the characteristics of opengl, when using this processing method, it is necessary to operate the image to be processed in a serial manner. This method processes ...

AI Technical Summary

Problems solved by technology

In related technologies, the image is usually processed based on GPU (Graphics Processing Unit, graphics processor) through opengl (OpenGraphics Library, which refers to a professional graphics program interface that defines a cross-programming language, cross-platform programming interface specification), However, due to the characteristics of opengl, when using this processing method, it is necessary to operate the image to be processed in a serial manner. This method processes the image in a single-threaded manner. Therefore, the speed will slow down when processing large-resolution images. There is a significant decline, and some image processing processes cannot guarantee real-time performance, thereby affecting the efficiency of real-time video transcoding; The speed of processing is proportional to the number of pixels in the image, so when processing the image through the CPU, if the image is processed to a higher resolution, the processing speed is not ideal, and the quality of image processing is not enough.

Images