AI identification system and method for article recovery and placement in industrial field

Abstract

The invention discloses an AI identification system and method for article recovery and placement in the industrial field. The AI identification system comprises a recovery box which is internally provided with a conveying device and a camera, the conveying device is arranged at the bottom in the recovery box, the camera is arranged above the conveying device so that at least part of the upper endface of the conveying device can be within the view field range of the camera; and the conveying device and the camera are electrically connected with a controller. According to the AI identificationsystem and method for article recovery and placement in the industrial field, numerous defects of a traditional scheme are overcome, a deep learning recognition scheme based on machine vision and corresponding physical structure design are provided, accurate classification recognition and recycling of articles in the industrial field are effectively achieved; and the specific article category canbe accurately identified, the problem of a factory in article recovery management is solved, the operation and management efficiency of the factory is obviously improved, and the waste is eliminated.

Description

technical field [0001] The present invention relates to the technical field of intelligent recycling of articles, in particular to an AI identification system and method for recycling and placing articles in the industrial field. Background technique [0002] In the factory, there are many scenes that need to put the used tools back and put them away, so that they can be used and searched efficiently next time. However, how to identify and manage whether the currently returned tool is a previously used tool is a prominent pain point in factory management. [0003] In the traditional method, there are many methods such as RFID, drop detection, weight detection, etc., but they all have very obvious problems. For many tools, such as cutting knives, they need to run at high speed, and they are small in size, so they cannot be pasted on them. RFID is used for identification, drop detection and weight detection. The problem is more intuitive. It can only determine whether it is p...

AI Technical Summary

Problems solved by technology

However, how to identify and manage whether the currently returned tool is a previously used tool is a prominent pain point in factory management

[0003] In the traditional method, there are many methods such as RFID, drop detection, weight detection, etc., but they all have very obvious problems. For many tools, such as cutting knives, they need to run at high speed, and they are small in size, so they cannot be pasted on them. RFID is used for identification, and the problems of drop detection and weight detection are more intuitive. It can only determine whether it is present, but cannot determine whether it is the same kind of tool. There are huge loopholes in supervision and it is easy to cause the loss of factory assets.

Images