Production line workpiece identification method, server and terminal

Abstract

The embodiments of the invention provide a production line workpiece identification method, a server and a terminal and belong to the production line visual inspection field. The production line workpiece identification method applied to the server includes the following steps that: workpiece information about a workpiece on a production line is received from the terminal, wherein the workpiece information includes a category identification code and a workpiece image; and a neural network corresponding to the category identification code in the workpiece information is adopted to identify theworkpiece image, so that an identification result for the workpiece is generated, wherein the neural network is uniquely corresponding to the category identification code, and is trained with the category identification code and the workpiece image corresponding to the category identification code adopted as a label. A big data analysis-based artificial intelligence technology is applied to the workpiece identification of the production line, so that the accuracy of the workpiece identification can be ensured, the identification rate of the workpiece is significantly improved, problems in judging whether a household electrical appliance is provided with screws and whether the label of the household electrical appliance is correct can be solved, and products of which some screws are not installed due to negligence, products having no labels, and products having incorrectly printed labels can be prevented from flowing into the market.

Description

technical field [0001] The invention relates to the field of visual detection of production lines, in particular to a method for identifying workpieces in a production line, a server and a terminal. Background technique [0002] With the continuous improvement of the integration of production automation, visual inspection is usually applied in the production, assembly or packaging of product lines. Visual inspection technology refers to the use of scientific and technological means to replace human eyes to measure and judge products produced on the production line. [0003] At present, the visual products used to process industrial images on the market all load traditional algorithms locally on the industrial computer, and then the industrial computer locally recognizes the collected industrial images, which is not conducive to the large-scale classification training application of images, and is not easy to identify The iterative upgrade of the algorithm leads to poor robu...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the embodiment of the present invention is to provide a production line workpiece recognition method, server and terminal, to at least solve the technical problem of poor robustness caused by industrial image recognition locally in the industrial computer in the prior art

Images