SOC (system-on-chip)-embedded machine vision equipment based on FPGA (field programmable gate array)

Abstract

The invention relates to the technical field of visual inspection for quality, and in particular to SOC (system-on-chip)-embedded machine vision equipment based on an FPGA (field programmable gate array). The SOC-embedded machine vision equipment based on the FPGA is characterized by comprising an SOC board card, a PC (personal computer), a power supply system and cameras, the power supply systemis connected with the SOC board card and the PC and used for supplying power to the same, the cameras are connected with the SOC board card and used for shooting video images, the SOC board card is connected with the PC and used for acquiring and processing the video images shot by the cameras and sending processed video image information to the PC, and the PC is connected with a local network, and is used for receiving and processing the video image information, displaying the video image information and controlling the SOC board card. The SOC-embedded machine vision equipment based on the FPGA is capable of acquiring and processing the video images shot by high-resolution or ultrahigh-resolution cameras rapidly.

Description

technical field [0001] The invention relates to the technical field of quality visual detection, in particular to a FPGA-based SOC embedded quality visual detection device. Background technique [0002] In the current domestic and international scope, quality visual inspection generally adopts the inspection mode of industrial computer. With the continuous improvement of factory production requirements, such as the improvement of production resolution, production speed, and production algorithm, PC industrial computers have been unable to meet these needs. At present, the PC industrial computer in the machine vision industry has the following problems: 1. The problem that high or ultra-high resolution cameras cannot be collected and processed; 2. Some image processing algorithms run a little slower on the industrial computer; 3. Some image processing algorithms The problem that it cannot run on the PC industrial computer. [0003] Popular machine vision hardware architectu...

AI Technical Summary

Problems solved by technology

With the continuous improvement of factory production requirements, such as the improvement of production resolution, production speed, and production algorithm, PC industrial computers have been unable to meet these needs.

At present, the PC industrial computer in the machine vision industry has the following problems: 1. The problem that high or ultra-high resolution cameras cannot be collected and processed; 2. Some image processing algorithms run a little slower on the industrial computer; 3. Some image processing algorithms The problem that it cannot run on the PC industrial computer

These are difficult or impossible to handle real-time video images such as 8K and 16K on industrial computers

A certain algorithm in the large window detection algorithm takes 200ms on the industrial computer, which means that it will occupy a very large resource overhead of the system, and the industrial computer cannot be used to do other tasks.

Moreover, the industrial computer is bulky and the product price is high, usually 5K-100K, or even more

[0004] Although there are very few foreign companies that have developed auxiliary equipment based on DSP+ARM according to the above-mentioned changes in actual needs, DSP or ARM is based on the instruction level, and the speed still has certain defects, so it cannot be well satisfied. The development needs of the above industries

Images