Human body size estimation method based on a radial basis function neural network

Abstract

The invention relates to a human body size estimation method based on a radial basis function neural network. The human body size estimation method comprises the following steps: collecting static human body net body data; preprocessing the collected human body net body data, and constructing a human body database; constructing a radial basis function neural network model of human body size estimation, an input layer of the radial basis function neural network model being a key part size which is easy to measure, and an output layer being a detail size which is not easy to measure; dividing the data in the human body database into a training sample set and a test sample set, training the radial basis function neural network by using the training sample set, randomly selecting a plurality of samples from the test sample set for simulation test after the training is completed, and estimating the detail size which is not easy to measure by the human body. The method can quickly and effectively calculate the size of the human body required by the garment making plate.

Description

technical field [0001] The invention relates to the technical field of human body size data collection, in particular to a method for estimating human body size based on a radial basis neural network. Background technique [0002] Accurate collection of human body data is the basis for the design and development of ideal clothing products, one of the key factors affecting clothing fit and consumer satisfaction, and one of the areas of focus and research in the clothing industry. With the rapid development of the Internet economy, the traditional manual measurement and non-contact 3D anthropometry technology is either insufficient in efficiency and accuracy, or is subject to high prices, and it is difficult to meet the new requirements of "remote, fast, convenient and accurate". need. At the same time, the details of the size required for some clothing structures are difficult to obtain and the accuracy is difficult to guarantee, such as the length of the upper crotch arc of...

AI Technical Summary

Problems solved by technology

In the field of clothing, some scholars have applied the perceptron and BP neural network models to human body size estimation, but these models have the disadvantages of slow convergence and easy to fall into local extreme values.

Images