CONVOLUTIONAL NEURAL NETWORK FOR CLASSIFICATION OF COTTON BY DEGREE OF OPENING
Abstract
This article describes the development of a cotton classification algorithm based on a convolutional neural network (CNN). The neural network consisted of convolutional layers, subsampling layers, and fully connected layers. The goal of this work was to classify cotton samples based on their degree of openness, while minimizing classification errors and achieving high accuracy. Data obtained by computer technology and image processing software were used in the algorithm creation process. The authors conducted a series of experiments with different CNN parameters and training sets to optimize the classification process. The final algorithm was tested on real cotton samples and demonstrated high classification accuracy. The results of this work confirm the effectiveness of using convolutional neural networks for cotton classification based on degree of openness. This article has practical value for controlling and regulating the technological parameters of a cotton-picking machine. It can also be useful for textile industry manufacturers, as well as for education in the fields of machine learning and artificial intelligence