基于灰联系规则的回转窑火焰图像检索方法.pdfVIP

第 39 卷第 5 期 中南大学学报自然科学版( ) Vol.39 No.5 2008 年 10 月 J. Cent. South Univ. (Science and Technology) Oct. 2008 基于灰关联规则的回转窑火焰图像检索方法 1 1 2 1 张红亮 ，李 劼 ，陈湘涛，邹 忠 (1. 中南大学 冶金科学与工程学院，湖南 长沙，410083； 2. 中南大学 信息科学与工程学院，湖南 长沙，410083) 摘 要：将数据挖掘中的灰关联分析引入基于内容的图像检索中，提出一种基于灰关联度的回转窑火焰图像的检 索方法。通过分析火焰图像特征值，并结合生产运行数据挖掘得到关联规则；应用灰关联度作为加权因子计算被 检索图像与数据库中图像的相似度，从而得到一系列相近检索结果；根据用户的相关反馈，查询得到更优结果； 设计和实现了检索系统的原型机，并应用从某氧化铝厂采集的图像和生产数据进行图像检索实验。实验结果表明： 该方法能够较快而有效地从图像数据库中检索得到较满意的结果。 关键词：回转窑；火焰图像；基于内容的图像检索；灰关联度；相似度 中图分类号：TP391 文献标识码：A 文章编号：1672−7207(2008)05−0881−06 Gray association rule based on rotary kiln flame image retrieval 1 1 2 1 ZHANG Hong-liang , LI Jie , CHEN Xiang-tao , ZOU Zhong (1. School of Metallurgical Science and Engineering, Central South University, Changsha 410083, China; 2. School of Information Science and Engineering, Central South University, Changsha 410083, China) Abstract: The association analysis of data mining was introduced into the content based image retrieval (CBIR), and a novel CBIR method was put forward. The features of the flame image were analyzed, from which the gray association rules were mined; the similarities between the queried images and the images in database were calculated based on the weight factors that were derived from the gray association rules, and as a consequence, several relevant flame images were retrieved; through the user’s relevance feedbacks, the final optimal results were achieved. A prototype machine was desi