Intelligent Recognition Of Garment Elements Based On Image Processing Technology

Chinese Shadow Play is a treasure of Chinese traditional culture with local characteristics,which contains rich folk awareness and cultural connotations.As an ancient folk art,the rich patterns in Shadow clothing contain unique aesthetic characteristics and traditional cultural connotation,which is the essence of human art creation.With the development of the times,the forms of entertainment are becoming more and more diversified.Shadow play and Shadow art are facing the crisis of withdrawing from the historical stage.In order to integrate it into the information society more effectively and spread more widely,the computer image processing technology is applied to the research of Shadow pattern in this paper.On the one hand,it can achieve accurate and objective identification and digital preservation.On the other hand,we can systematically carry out a comparative study of different roles and different regions to highlight the characteristics of Shadow culture.It is helpful to make full use of the elements of traditional culture for better redesign and innovative application,activate the vitality of traditional culture,and realize the protection and inheritance under the new situation.Therefore,this paper makes the following research.First of all,in order to identify the local pattern contour objectively and effectively,the selective target extraction is realized.Aiming at the characteristics of rich local details,high color saturation and strong interference of background information in Shadow images,an intelligent algorithm GABRS(Grab Cut Algorithm Based on RTV and SLIC)for extracting Shadow pattern contours is designed: 1)the relative total variation model is used to separate the noise from the main structure to realize the smooth processing of the image;2)the Grab Cut algorithm is designed,and the optimal number of super-pixels is determined by analyzing the boundary compactness index of the pattern outline,so as to realize the optimal segmentation of the local pattern;3)the contour of the segmented Shadow pattern is extracted by using Canny operator.Secondly,according to the color composition and pattern characteristics of Shadow clothing,a method of recognizing the main color of Shadow pattern based on color clustering is designed: 1)the automatic extraction of Shadow pattern is realized by texture filtering and image segmentation;2)the DPC clustering is used to cluster the single Shadow pattern for the first time to extract the main color of each pattern.3)according to the different roles,the extracted dominant colors are clustered again,and the dominant colors of each kind of roles are obtained.On this basis,combined with the production technology of Shadow and the characteristics of the characters,the distribution characteristics of hue,saturation and lightness of the main color are compared and analyzed.It visually shows the distribution of the main color from a scientific point of view,which provides a color matching reference for the follow-up pattern redesign based on Shadow culture factors.Finally,the color and pattern features extracted from the Shadow clothing pattern research are taken as the design basis,and the contours and color factors that can reflect the connotation of Shadow culture are selected.Combined with the combination form and composition characteristics of Shadow pattern,the selected outline pattern is designed and evolved based on shape grammar,the reconstructed pattern style is analyzed,and the corresponding color factor is selected for pattern filling.The redesigned reconstructed pattern is applied to modern clothing design,and finally it is displayed virtually in 3D software.