TEXTILE DYEING PROCESS AND DYEING RECIPE PREDICTION USING ARTIFICAL INTELLIGENCE

Abstract

In this study, a comprehensive analysis is presented on the determination and estimation of the sample color or target color taken from the customer in the laboratory department of textile dyeing companies. The importance of the textile industry in the world and Turkey is also mentioned. In the report published by Statista, it is seen that the textile industry has a share of 3.3% in Turkey and the world. In this work, a sample study was conducted in a textile finishing company and the processes were shared. First, the classical processes used to determine the target color are explained in detail. Then, it was mentioned how the data obtained with the spectrophotometer device is used in color estimation using machine learning methods and artificial neural networks According to the results of the examination, it is seen that the data obtained with the hyperspectral camera device is estimated by the long short-term memory (LSTM) model, since the spectrophotometer device is expensive and does not give accurate results recently. In addition, it has been observed that this model gives better results than the same model created from the data obtained with the spectrophotometer device