Product defect detection equipment and method

Abstract

The invention relates to a product defect detection method and equipment. The product defect detection method comprises the following steps: according to obtained sample data of a model construction participant, training a data model according to a data model training instruction to obtain an intermediate parameter, and encrypting the intermediate parameter; decrypting the encrypted intermediate parameters, and fusing the decrypted intermediate parameters to obtain fused parameters; and then, updating model parameters of the data model according to the fusion parameters, and obtaining a product defect detection classification model with industrial product data as input and a product surface defect detection result as output. The problem that data in the age of big data are not easy to share is solved, the data safety is further protected, meanwhile, defect data are richer, and the model is more accurate.

Description

technical field [0001] The invention relates to the field of product defect detection, in particular to an MES-oriented product defect detection device and method based on federated learning. Background technique [0002] The manufacturing execution system (MES) is designed to help major companies realize detailed process scheduling, production unit allocation, manufacturing resource allocation and status reporting, document control, product tracking and product list management, data collection, product quality management, Workshop inventory management, etc., are conducive to solving the visualization and controllability of the factory production process, and improving the level and ability of manufacturing execution. Among them, product surface defect detection is an important part of product quality management. [0003] In the context of the industrial Internet, artificial intelligence is developing rapidly with a large amount of labeled data. When performing defect detec...

AI Technical Summary

Problems solved by technology

When performing defect detection on the data-driven MES-oriented product surface, there are poor data quality in the existing technology (more junk data, missing some data labels), insufficient database samples, data isolation and data dispersion caused by data islands, data Lack of shared security, high cost of data privacy protection, etc.

Images