System of product defect discrimination using an unsupervised learning based regional convolutional auto-encoder and Method of product defect discrimination using an unsupervised learning based regional convolutional auto-encoder

摘要

The present invention relates to a system for discriminating a defect of a product using an unsupervised learning convolutional auto-encoder by defining an image and determining the defined image by region, and a method for discriminating the defect of the product using unsupervised learning based on a regional convolutional auto-encoder. The system of the present invention includes an image aligning unit for aligning a first product image provided from the outside according to a standard of the product; a regional convolutional auto-encoder receiving the first product image aligned by the image aligning unit; and a discrimination reference determining unit connected to the regional convolutional auto-encoder.