Rapid detection method for spandex content in cotton, polyester and spandex blended fabrics

Abstract

The invention discloses a method for rapidly detecting the content of spandex in a blended fabric of cotton, polyester, and spandex. The method comprises the following steps: step one, measuring the spandex contents of each blended fabric in a calibration set and a prediction set; step two, obtaining the near infrared spectral data of each blended fabric, and subjecting all near infrared spectrumsof the calibration set and the prediction set to a normalization pretreatment; step three, dividing the sets into subsets by a self-organized neural network model; step four, establishing corresponding subunit prediction models through partial least squares according to the spandex content corresponding to each subunit in the step one; step five, choosing the subunit prediction model having the highest prediction accuracy; and step six, through the subunit prediction model corresponding to the subunit, detecting the spandex content of a blended fabric to be detected. The provided method can effectively improve detection accuracy and speed, and protects the environment and health of people.

Description

technical field [0001] The invention relates to the field of spandex detection, in particular to a rapid detection method for spandex content in cotton, polyester and spandex blended fabrics. Background technique [0002] Self-organizing map neural network, namely Self Organizing Maps (SOM), can perform unsupervised learning clustering on data. Its idea is essentially a neural network with only the input layer-the hidden layer (mapping layer). A node in the hidden layer represents a class that needs to be clustered. The "competitive learning" method is adopted during training, and each input sample finds a node that best matches it in the hidden layer, which is called its activation node. Then, the parameters of the activation node are updated with the stochastic gradient descent method, and at the same time, the parameters of the points adjacent to the activation node are also updated appropriately according to their distance from the activation node. [0003] The conten...

AI Technical Summary

Problems solved by technology

At present, the quantitative analysis of spandex content is mainly based on chemical dissolution method, which is time-consuming and labor-intensive, and consumes a large amount of chemical reagents, which will endanger the environment and human health

[0004] Another detection method is near-infrared spectroscopy, which can more accurately detect the fiber content of blended fabrics, but the content of spandex in cotton, polyester and spandex blended fabrics is generally low, and is easily affected by factors such as fabric dyes and structures , resulting in a large relative error

Images