Open AccessRealization of clothing image contour extraction and collar segmentation
Author(s) -
Ling Zhang,
Chao Wang,
Yanhong Zhang
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1790/1/012091
Subject(s) - clothing , collar , preprocessor , computer science , image processing , artificial intelligence , computer vision , image segmentation , segmentation , digital image processing , sample (material) , edge detection , software , realization (probability) , image (mathematics) , engineering , mathematics , geography , mechanical engineering , chemistry , statistics , archaeology , chromatography , programming language
With the change of personalized ideas, the clothing and apparel industry is developing towards small batches and multiple varieties. Looking for fast and efficient apparel image processing and recognition is the most urgent problem to be solved by apparel companies. The establishment of a collar style sample library is the prerequisite for apparel image processing and recognition research. Apparel image preprocessing is the basis of research. This article takes the clothing style map as the research object, establishes a collar style sample library, and takes a round neck T-shirt image as an example, which contains 8 types of collars commonly used in clothing, and each collar type corresponds to 60 images. The advantages and disadvantages of commonly used image graying, sharpening, edge detection, morphological processing, and image segmentation processing methods are compared and analyzed. Corresponding image preprocessing schemes have been formulated to provide new ideas for automatic pattern recognition, software development and application.