Scrap Sorting via Color and Spectral Data Vectors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current scrap sorting technologies face challenges in accurately and efficiently classifying and sorting scrap materials into specific categories, such as metals and plastics, due to variations in shape, size, and properties, which affects their reuse and reduces pollution compared to refining virgin materials.
Innovation Solution
A system utilizing a vision system to image scrap particles on a conveyor, combined with laser spectroscopy, generates color and spectral data vectors to classify materials into predefined categories using machine learning algorithms, enabling precise sorting and separation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple sorting techniques (hand sorting, air sorting, vibratory sorting, color based sorting, magnetic sorting, spectroscopic sorting) are used on a conveyor belt system, then sorting capability is improved, but device complexity increases
Solution Approach 1:
The patent combines color-based sorting and spectroscopic sorting into a single integrated system. The vision system captures both color information and spectroscopic data simultaneously, merging multiple sorting capabilities into one unified apparatus that processes materials through a single conveyor belt system rather than requiring separate sorting lines.
Solution Approach 2:
The sorting system is designed to handle multiple material types (metals, plastics, and other materials) using the same conveyor belt and detection infrastructure. The system can classify materials into different categories (ferrous metals, non-ferrous metals, plastics, etc.) using a universal platform that adapts its classification algorithms based on the material being processed.
2Measurement precision
If spectroscopic analysis is performed on each particle, then material classification accuracy is improved, but processing speed decreases
Solution Approach 1:
The vision system performs preliminary color-based classification and particle identification before spectroscopic analysis. This preliminary action filters and pre-sorts particles, allowing the spectroscopic system to focus only on particles that require detailed spectral analysis, thereby maintaining high processing speed while achieving accurate material classification.
Solution Approach 2:
The system applies spectroscopic analysis selectively to specific particle types or size ranges that benefit most from spectral characterization. Not every particle undergoes full spectroscopic examination; instead, the system uses color-based sorting for preliminary classification and applies spectroscopic methods only where needed to achieve the required classification accuracy.
3Adaptability or versatility
If hand sorting by operators is used, then flexibility in classification is improved, but productivity decreases
Solution Approach 1:
The patent replaces manual hand sorting with an automated vision system that uses color imaging and spectroscopic analysis to classify materials. The mechanical and human operations of inspection and classification are substituted with optical detection systems and automated decision-making algorithms, eliminating the need for human operators while increasing sorting speed and consistency.
Solution Approach 2:
The system uses multiple detection parameters (color values, spectral signatures, particle size, shape characteristics) to classify materials automatically. By changing from single-parameter manual classification to multi-parameter automated detection, the system achieves both the flexibility of human judgment and the high-speed processing capability of automated systems.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system achieves high accuracy in sorting scrap materials into specific categories, enhancing their reuse and reducing environmental impact by improving the efficiency of material recycling and reducing pollution.
Implementation Method 1
A moving conveyor containing scrap particles is imaged using a vision system to create an image corresponding to a timed location of the conveyor
Implementation Method 2
A light beam is generated and directed to the particle on the conveyor downstream of the vision system and within a target area using a scanner assembly including a beam deflector
Implementation Method 3
At least one emitted band of light from the particle in the target area is isolated and measured at a selected frequency band using a detector to provide spectral data for the particle
Data Source
Figure 1
Figure 2
Figure 3~8
AI summary
A system and a method of sorting scrap particles includes imaging a moving conveyor containing scrap particles using a vision system to create an image. A computer analyzes the image as a matrix of cells, identifies cells in the matrix containing a particle, and calculates a color input for the particle from a color model by determining color components for each cell associated with the particle. A light beam is directed to the particle on the conveyor downstream of the vision system, and at least one emitted band of light from the particle is isolated and detected at a selected frequency band to provide spectral data for the particle. The computer generates a data vector for the particle containing the color input and the spectral data, and classifies the particle into one of at least two classifications of a material as a function of the vector.