Garment texture recognition and classification method based on LBP and GLCM

Abstract

The invention relates to a garment texture recognition and classification method based on LBP and GLCM. The classification method for the texture types in fashionable garment images comprises the following steps: S1, extracting LBP texture features; S2, designing a GLCM texture image feature statistic extraction function; S3, calculating a texture feature value; and S4, training the feature vectors by using an SVM algorithm, and classifying the texture pictures. The method has the advantages that the challenge of texture recognition in fashionable clothes is well solved in the actual production process, the constraint of manual recognition is liberated, convenience is brought to later data analysis and algorithm development optimization, and the method has important practical significance.

Description

technical field [0001] The invention relates to the technical field of data recognition, in particular to a clothing texture recognition and classification method based on LBP and GLCM. Background technique [0002] Content-based image retrieval (CBIR) generally includes four stages: image acquisition, feature extraction, image classification, and image retrieval. CBIR has two core issues: how to achieve fast and effective image classification and retrieval, the key lies in the selection of which algorithm to extract which features; how to establish an effective image classification and recognition system, the key lies in the selection of classification algorithms. The fashion database comes from the Internet and can provide customers with high-quality retrieval of fashion styles and fashion elements. However, there are many texture patterns in natural images, and the texture structures used in clothing design are also ever-changing. It is conceivable that the recognition of...

AI Technical Summary

Problems solved by technology

[0010] The above patent document: CN201510347067.7, a texture image classification method based on BoF and multi-feature fusion, effectively overcomes the shortcomings of GGCM for the low accuracy rate of large texture classification, and at the same time makes up for the weakness of BoF feature space information loss. A more accurate and robust texture image classification method; Patent Document: CN201510513760.7 An airport target automatic recognition method based on line classification and texture classification effectively avoids the error of airport target extraction caused by purely relying on line feature classifiers In this case, the accuracy of automatic airport target recognition can be effectively improved; but for a feature extraction of a small number of representative texture images, the feature representation of some statistics is obtained, and then the SVM classification model is trained. According to the obtained The classification method of the model to predict our large number of product maps and obtain the retrieved texture results has not yet been reported.

Images