Optical image processing system based on embedded GPU

Abstract

The invention discloses an optical image processing system based on an embedded GPU. The system comprises a master embedded GPU module, a slave embedded GPU module, an FPGA module and a storage module, wherein the storage module is connected with the main embedded GPU module, and an operating system is stored in the storage module and used for installing the operating system for the main embeddedGPU module; and the FPGA module is used for receiving the configuration parameters and the control commands sent by the main embedded GPU module, sending acquired image data to the slave embedded GPUmodule for image data processing, receiving a processing result of the slave embedded GPU module and outputting the processing result to the main embedded GPU module. According to the invention, system structure simplification, integration, miniaturization, high flexibility and low cost can be realized.

Description

technical field [0001] The invention belongs to the field of electronic technology, and more specifically relates to an optical image processing system based on an embedded GPU. Background technique [0002] Display screens such as LCDs are widely used in industries such as computers, communications, and consumer electronics. The manufacturing process of the display screen is relatively complicated, which may lead to defects in the display screen, so automatic optical inspection is required, namely AOI (Automated Optical Inspection). An important part of the AOI system is the image processing system. AOI's image processing system will need to perform a lot of image computing and processing work such as deep learning. With the development of the display industry of LCD and OLED, the complexity and requirements of the image processing system used for AOI have also increased exponentially, and the algorithm workload involved in image processing is heavy. [0003] The existin...

AI Technical Summary

Problems solved by technology

[0004] (1) The system has limited computing and processing capacity, slow response speed, high overall system cost, and large system volume;

[0005] (2) The system structure is complex, and generally requires a combination of multiple functional modules. For example, for screens with different resolutions, different image acquisition units need to be configured;

[0006] (3) Separate system architecture, low flexibility in use and function

Images