2018/03/12 by Shervan Fekri-Ershad, Fekri-Ershad, Shervan
Computer Science · Engineering · #Advanced Image and Video Retrieval Techniques #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Image Retrieval and Classification Techniques #Industrial Vision Systems and Defect Detection
paper · pdf · doi:10.48550/arxiv.1803.04125
openalex publication_date 2018/03/12 · openalex created_date 2022/10/03 · openalex updated_date 2026/07/28
Texture analysis and classification are some of the problems which have been\npaid much attention by image processing scientists since late 80s. If texture\nanalysis is done accurately, it can be used in many cases such as object\ntracking, visual pattern recognition, and face recognition.Since now, so many\nmethods are offered to solve this problem. Against their technical differences,\nall of them used same popular databases to evaluate their performance such\nasBrodatz or Outex, which may be made their performance biased on these\ndatabases. In this paper, an approach is proposed to collect more efficient\ndatabases of texture images. The proposed approach is included two stages. The\nfirst one is developing feature representation based on gray tone difference\nmatrixes and local binary patterns features and the next one is consisted an\ninnovative algorithm which is based on K-means clustering to collect images\nbased on evaluated features. In order to evaluate the performance of the\nproposed approach, a texture database is collected and fisher rate is computed\nfor collected one and well known databases. Also, texture classification is\nevaluated based on offered feature extraction and the accuracy is compared by\nsome state of the art texture classification methods.\n