Abstract:

Objective In the context of intelligent and automated clothing structure design, there are many problems in the quantitative representation of fold morphology, such as a wide variety and a lot influencing factors. In order to solve this problem, this paper proposes a parametric characterization method for clothing fold morphologies, and establishes the mapping relationship between its two-dimensional plane style and three-dimensional structure, so as to realize the automatic generation of fold personalized patterns.

Method The parametric theory was used and a series of parameter values were set. Relevant parameters were adopted to constrain and describe the structural dimensions of geometry. By combining the structural drawing principle of folds and the relationship between internal fold structures, 14 characteristic parameters that can quantify the fold morphology were proposed. Five skirts of the same fabric were prepared as experimental samples, including a total of six common pleated styles. There were 22 sets of experiments designed from these 5 samples. In the experiment, physical photos of clothing folds were taken under the same angle of view and the same plane. Then the photos were subjected to image processing for feature parameter extraction.

Results From the quantification results of the fold images of specimen A-E, it can be seen that each fold in the sample can be decomposed into multiple single folds of different sizes and orientations to characterize. These folds can be quantified with the extracted 14 characteristic parameters and 11 representative folds were extracted from five specimens. Estimates of the parameters in these folds were found. Then the data were compared with their actual measurements. SPSS were utilized to pair estimates and measurements for different types of folds. The results show that the folds, which quantified by this method, has a significant correlation between the estimated and measured values used for testing. At the same time, no significant difference exists between the estimated and measured values of the folds at the 95% confidence interval, indicating that the folds estimated and measured values are relatively close, and this method has a certain accuracy.

Conclusion After studying the quantification ability of different characteristic parameters on folds and conducting related experiments, the results show that each characteristic parameter extracted in this paper has practical application value. At the same time, this method has high accuracy. No significant difference exists between the experimental verification data and the data read by the tester. This method is more systematic. It quantifies and characterizes various factors affecting fold morphology through mathematical means, which can realize the systematic and non-subjective evaluation of fold morphological characteristics. This method is flexible and has good quantification capabilities for a variety of different types of folds. The fold parameterization method proposed in this paper can effectively realize the quantitative characterization of clothing fold morphology. It has practical application and feasibility in the parameter acquisition and style recognition of clothing intelligent plate making, which provides a new research idea for the intelligent development of clothing engineering.

Key words: clothing fold, fold morphology, parameterization, structural design, intelligent plate making, clothing engineering