Coin Recovery Vision Sorting for Automotive Scrap Residues
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing recycling technologies struggle to efficiently recover valuable materials such as monetary coins, jewelry, and PCBs from automotive shredder residues, which are often mixed with other scrap materials, and require costly laser-induced breakdown spectroscopy analysis.
Innovation Solution
A machine learning-based vision system is employed to identify and sort valuable scrap pieces, including monetary coins, jewelry, and PCBs, using a conveyor system, cameras, and automated sorting devices, without the need for laser-induced breakdown spectroscopy, by capturing images and applying machine learning algorithms to classify and sort materials based on physical characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If laser-induced breakdown spectroscopy analysis is used to identify and sort valuable scrap pieces, then measurement precision is improved, but device complexity and cost increase significantly
Solution Approach 1:
The patent replaces the complex laser-induced breakdown spectroscopy system with a simpler machine learning-based vision system using cameras and image processing. This substitution maintains identification accuracy while dramatically reducing device complexity and cost by using optical imaging and computational algorithms instead of laser plasma generation and spectral analysis equipment
Solution Approach 2:
The patent creates visual copies (images) of scrap pieces using cameras and processes these copies through machine learning algorithms to identify valuable materials. This approach avoids the need for direct physical interaction with the scrap pieces through laser ablation, simplifying the system while maintaining measurement precision through digital image analysis
2Productivity
If manual sorting of valuable scrap pieces is performed, then device complexity is reduced, but productivity and labor costs decrease
Solution Approach 1:
The patent implements an automated sorting system where the machine learning vision system independently identifies, classifies, and directs valuable scrap pieces without human intervention. The system serves itself by automatically making sorting decisions and controlling the sorting mechanism, thereby increasing productivity while managing complexity through integrated automation
Solution Approach 2:
The patent uses strong computational algorithms and machine learning models as intellectual 'oxidants' to rapidly process and analyze visual data, accelerating the identification and sorting process. This computational acceleration enables high-speed automated sorting that maintains simplicity while dramatically increasing throughput compared to manual methods
3Productivity
If valuable scrap pieces are not recovered from automotive shredder residues, then processing speed is improved, but loss of valuable materials increases
Solution Approach 1:
The patent applies preliminary identification and classification of valuable scrap pieces during the recycling process using the vision system, before final processing occurs. This preliminary action ensures that valuable materials are identified and separated in advance, preventing loss while maintaining overall processing speed through early detection and sorting
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
A material sorting system sorts materials utilizing a vision system that implements a machine learning system in order to identify or classify each of the materials, which are then sorted into separate groups based on such an identification or classification determining that the materials have a specified geometric shape. Such a system can sort monetary coins or other valuable metals from other forms of scrap.