A metal identification method based on information entropy

Abstract

The invention discloses a metal identification method based on information entropy, which belongs to the technical field of image identification. In this method, multilevel wavelet transform is firstly performed on the sample image, and then K-means algorithm is used for primary recognition. For unrecognized samples, the transformed multilevel wavelet transform entropy value is analyzed, and decision tree algorithm is used for secondary recognition. This method ensures the accuracy of the algorithm while improving the recognition speed through multi-level recognition. The algorithm has fast speed, high accuracy and high recognition efficiency, which can meet the real-time requirements. The method supports parallel processing, the equipment required for identification is simple and environmentally friendly, and has wide applicability. To a certain extent, the method of the present invention can distinguish more types of non-ferrous metals.

Description

technical field [0001] The invention belongs to the technical field of image recognition, and specifically refers to a fast metal recognition method based on information entropy, which is suitable for the recognition of metal sorting pipelines. technical background [0002] With the development of the economy, the Chinese market has become increasingly dissatisfied with the output and high energy consumption of primary metals. At the same time, recycled metals have received more and more attention. In recent years, the amount of scrap metal in my country has become increasingly large, and the demand for recycling non-ferrous metals has grown rapidly, and the recycling and dismantling industry will also usher in a period of rapid development. Taking automobiles, which are the main recycling objects of non-ferrous metals, as an example, standardizing the development of scrapped car recycling and dismantling industry is crucial to the development of my country's auto industry a...

AI Technical Summary

Problems solved by technology

Taking automobiles, which are the main recycling objects of non-ferrous metals, as an example, standardizing the development of scrapped automobile recycling and dismantling industry is crucial to the development of my country's automobile industry and circular economy, but the current domestic recycling rate of scrapped automobiles is less than 20%

This method has the following problems: (1) toxic gas is easy to escape and volatilize, and some metals are melted in the slag to cause losses; (2) a large amount of non-metallic components such as plastics cannot be recovered

[0011] It can be seen from the above technologies that traditional hydrometallurgy and pyroprocessing have great damage to the environment, and their efficiency is low, which cannot meet the real-time requirements of assembly line operations.

The sorting effect of eddy current is affected by the size and type of metal objects. The strength of eddy current magnetic field is difficult to adjust in time according to the metal fragments on the fast-moving production line. It is generally only used to generally distinguish plastics and non-ferrous metals.

Coupled with the huge energy consumption and noise caused by the magnetic field, the eddy current screening method is not suitable for the recycling of scrap metal in the new era

The use of flotation agents also consumes resources, and the real-time requirements are difficult to meet

In X-ray identification, the commonly used portable X-ray machines and security inspection channels cannot meet the real-time and accuracy requirements respectively, and cannot be applied on conveyor belts with high processing speed requirements.

Images