An online recognition method of seed cotton mulch film based on hyperspectral imaging and deep learning

Abstract

The invention discloses an on-line identification method of seed cotton mulch based on hyperspectral imaging and deep learning. The hyperspectral imager is used to obtain the reflection spectrum image of seed cotton mulch, and a deep learning network composed of a stacking weight autoencoder and a particle swarm optimized extreme learning machine is constructed. For online recognition of hyperspectral images, the present invention uses a network composed of stacked weighted autoencoders and extreme learning machines in deep learning to classify hyperspectral images of seed cotton mulch, and introduces a weighting mechanism in each layer of autoencoders to ensure multiple The channel input advantage reduces the impact of noise; the weights and biases of the extreme learning machine are randomly determined, which is prone to overfitting. The particle swarm optimization algorithm is used to optimize the weights and biases of the extreme learning machine to ensure the recognition speed At the same time, the classification accuracy is improved. The deep learning network composed of stacked weighted autoencoder and extreme learning machine can be used for online recognition of seed cotton mulch.

Description

technical field [0001] The invention belongs to the technical field of foreign fiber identification of seed cotton, and in particular relates to an online identification method of mulch film of seed cotton based on hyperspectral imaging and deep learning. Background technique [0002] my country is a big country of cotton production and consumption, and cotton processing and weaving play an important role in the national economy. As a major cotton-producing province in my country, Xinjiang has widely used plastic film mulching technology in cotton planting, and cotton picking production is highly mechanized. During the mechanical picking process, seed cotton is mixed with a large amount of plastic film. If it is not cleaned thoroughly, it will follow the processing link. If it enters the lint, it will definitely affect the quality of the textile and the dyeing quality of the textile. At present, the mulch fragments contained in machine-picked cotton have become the fundament...

AI Technical Summary

Problems solved by technology

As a major cotton-producing province in my country, Xinjiang has widely used plastic film mulching technology in cotton planting, and cotton picking production is highly mechanized. During the mechanical picking process, seed cotton is mixed with a large amount of plastic film. If it is not cleaned thoroughly, it will follow the processing link. Into the lint, it will definitely affect the quality of the textile and the dyeing quality of the textile

Images