abstract |
A vision inspection system and method using distance learning of product defect images are disclosed. The vision inspection system using distance learning of product defect images comprises: an edge platform including a defect inspection module, a defect determination module, and a manufacturing data transfer module for learning, wherein the defect inspection module is connected to a camera, a sensor, an LED light, and a controller, is equipped with a machine vision inspection system having a machine vision image analysis SW, reads a product ID, and provides a shape determination inspection, foreign substance inspection, and scratch inspection of camera image data of a product; middleware connected to the edge platform and interworking with a service platform; and the service platform which interworks with the edge platform through the middleware, provides defect prediction manufacturing intelligence, detects an atypical defective image through comparison with accumulated and stored defective image training data, and provides a foreign substance inspection of the camera image data of the product, shape inspection, and vision inspection through remote learning of product defect images that uses a deep learning algorithm which provides product normal/defective determination result data. |