Scrap plastic spectrum identification method based on XGBoost model

Abstract

The invention discloses a waste plastic spectrum identification method based on an XGBoost model. The method comprises the following steps: acquiring near-infrared reflectivity spectrum data of various plastics by using a near-infrared spectrometer; pCA analysis is carried out on the original spectral data, and the data dimension is reduced; an XGBoost algorithm is used to establish a waste plastic algorithm identification model; optimizing the performance of the recognition model by adopting grid search and K-fold cross validation, and finally establishing the recognition model; and based on the trained identification model, extracting the near infrared spectrum of the to-be-identified plastic sample, and identifying the near infrared reflection spectrum of the to-be-identified plastic sample according to the identification model to obtain an identification result of the to-be-identified plastic sample. According to the method, the data processing time is shortened, meanwhile, the accuracy of identifying various common plastics is improved, and application in actual production is facilitated.

Description

technical field [0001] The invention relates to the field of identification of waste and miscellaneous plastics, in particular to a spectral identification method of waste and miscellaneous plastics based on an XGBoost model. Background technique [0002] In the past few decades, the great progress of science and technology and the rapid development of economy and society have led to a huge increase in the demand and output of various industrial products. After decades of rapid development, my country's plastics industry has expanded its scale year by year, and its production capacity and output have continued to increase. At present, it has become an equally important basic industry with cement, wood and steel. [0003] Because of its light weight, non-corrosion, stable chemical properties, and low price, plastics are widely used in various fields such as electronics, automobiles, construction, and agriculture. They can be seen everywhere from clothing, food, housing, trans...

AI Technical Summary

Problems solved by technology

Improve the recycling rate of plastic products, and better solve the problem of low accuracy of traditional chemical and physical methods for classifying various plastics

Images