A method for fabric scanning pattern recognition based on multi-feature fusion

Abstract

The invention discloses a multi-feature fusion fabric scanning pattern recognition method, which first filters out the yarn texture of the fabric scanning image through a texture suppression fast smoothing filter algorithm, and performs grayscale conversion; then extracts the main color autocorrelation histogram respectively , edge gradient direction histogram, MSER feature and gray level co-occurrence matrix feature, to establish a sample image feature library; finally, the similarity between the sample images about these four types of features is used as a training sample, and a classifier is established through the AdaBoost algorithm to realize pattern recognition . It can be seen that the pattern recognition method of the present invention can automatically adjust the weights of various features and improve fabric pattern recognition rate.

Description

technical field [0001] The invention belongs to the technical field of digital image processing, and in particular relates to a multi-feature fusion fabric scanning pattern recognition method. Background technique [0002] With the improvement of the automation of textile production and design, the management of textile design resources by textile enterprises has changed from traditional manual management to automatic management. Many textile enterprises have accumulated a large number of fabric pictures of different patterns. In order to avoid repeated design and effectively utilize existing resources, an effective method for fabric pattern recognition is urgently needed. [0003] The computer pattern recognition process can generally be divided into three steps: preprocessing, feature extraction, and algorithm recognition. Through image preprocessing, image noise is reduced; then pattern features are extracted and a feature database is established; finally, according to t...

AI Technical Summary

Problems solved by technology

Bunte et al. proposed an image recognition technology based on color features in a paper titled Learning effective color features for content based image retrieval indermatology (Pattern Recognition, 2011,44(9):1892-1902); Kekre et al. titled Image Retrieval using Texture Features extracted from GLCM, LBG and KPE (International Journal of Computer Theory and Engineering, 2010, 2(5): 1793-8201) proposed an image recognition technology based on texture features; Kekre et al. titled Image Retrieval with Shape Features Extracted using GradientOperators and Slope Magnitude Technique with BTC (International Journal of Computer Applications, 2010, 6(8): 28-33.) The retrieval technology based on shape features proposed in the article; these single features only reflect the characteristics of a certain aspect of the pattern, and cannot Fully characterize the pattern features, pattern recognition using only a certain feature is often not accurate enough

Images