基于灰关联规则的回转窑火焰图像检索方法

Gray association rule based on rotary kiln flame image retrieval

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。

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 designed, and the image retrieval experiments were carried out based on the images and production data sampled from an alumina plant. The results show that the proposed method can achieve the satisfactory results efficiently and promptly.