Machine vision detection system and method

Abstract

An embodiment of the invention discloses a machine vision detection system and method. The machine vision detection system comprises at least two industrial cameras, an edge computing cloud and mobile equipment, wherein the at least two industrial cameras are arranged on the mobile equipment; the mobile equipment is used for driving the industrial cameras to move, so that each industrial camera acquires images of the to-be-detected object at different angles; and the edge computing cloud is used for receiving the images uploaded by the industrial cameras, processing the images and acquiring a detection result of the images. According to the machine vision detection system and method of the embodiment of the invention, image processing of the edge computing cloud can be shared, the use of an industrial personal computer is canceled, the light weight of the equipment end is realized, the equipment installation and debugging process is simplified, the use is flexible, the maintenance is convenient, and the cost is saved.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of machine vision, and in particular to a machine vision inspection system and method. Background technique [0002] Machine vision detection and recognition technology has been widely used in industrial production, such as visual inspection of printed circuit boards, automatic flaw detection of steel plate surface, measurement of parallelism and perpendicularity of large workpieces, detection of container volume or impurities, automatic recognition and classification of mechanical parts and geometry Size measurement etc. In addition, machine vision can be used effectively in many situations that are difficult to detect by other methods. [0003] Machine vision systems in the prior art generally install an industrial host on multiple stations, and the industrial host can be connected to multiple industrial cameras through cables, and the industrial cameras on each station can reali...

AI Technical Summary

Problems solved by technology

[0004] However, in the prior art, the industrial host on each station is equivalent to a "stand-alone version", and adding more stations requires adding an industrial host, and the investment cost of the enterprise is high, and the installation and debugging of the industrial host is time-consuming; the machines at each station Visual inspection is independent, and the inspection is only applied to the local station, and the inspection data cannot be reused; cable connection is used, which consumes cost, time, manpower and material resources for maintenance, and the transmission distance is short, which limits the flexibility of production; The algorithm upgrade needs to be operated separately, which is time-consuming and labor-intensive

Images