CSG News

Hyperspectral imaging assists metal sorting

FROM:- DATE:2018-08-13 16:37:43 CLICK:1258  

Vehicles, as well as electrical and electronic equipment, are often discarded as waste because it may be impractical to recycle this material in a cost effective manner. In the past, this recycling was done by manual sorting, a labor-intensive, time-consuming and inaccurate process. Using an automated system, 40%-50% steel and 30% other metals from recycled materials can be sorted by drum or inductive or vortex sorter, respectively. One of the biggest challenges in developing more complex systems is the sorting of non-ferrous metals such as aluminum, copper, brass, lead, stainless steel, silver and gold.

In the past, the recycling industry has been slow to accept automated methods for sorting such metal scrap. However, with the advent of hyperspectral cameras and robot-based picking and sorting methods, these processes can now be automated, freeing workers from heavy work and improving the efficiency of sorting metal alloys. As automated processes become more efficient, scrap suppliers can produce more recycled materials to increase profitability.

To achieve these goals, CSG, originally dedicated to mineral characterization and processing, has developed a robot-based system that includes multiple sensors, cameras, programmable logic controllers (PLCs), and robotic systems. “The classification of scrap metal sheets is a challenging problem due to the reflection of metal surfaces,” said the research team's research engineer. “In fact, the metal's reflection curve usually exhibits monotonous characteristics, and may have more plastics. The situation of multi-spectral shapes is different."

In order to perform the task of sorting non-ferrous metals, it is first necessary to locate the material and then classify it according to its spectral characteristics. These spectral features can be used to identify materials. In the visible and near-infrared (VNIR) spectrum, aluminum, copper, zinc, stainless steel, and brass all exhibit different spectral response characteristics. It can be seen that brass and stainless steel are the most difficult categories to distinguish because their spectral reflectance characteristics are very similar. Therefore, hyperspectral imaging systems must be used to distinguish each individual material.

—— - ——
CSG Copyright 2000-2018 / Article by COLOR SORTER