摘要
图像主题爬虫能获取网上特定主题的大量图像信息,对专业搜索引擎及数据挖掘应用都具有重大价值。针对目前基于图像内容检索主题爬虫的不足,提出了一种图像主题爬虫的设计方法,设计了一种新的爬虫系统框架,采用了基于颜色累加直方图的方法进行图像的特征提取与特征匹配。最后提出了优化爬虫的方法,改进爬虫的搜索策略,提高了爬虫的搜索效率。
The theme crawler of content based image retrieval (CBIR) can fetch large quantities of domain resources from the Web. It is of great value in both professional search engines and data mining companies. Due to the lack of theme crawler CBIR,a design of theme crawler CBIR is presented,and a frame of the crawler is given. Using color-based accumulate histogram to extract the feature of the image and match the feature. A method to improve the performance of the crawler is presented in the end.
出处
《广西师范大学学报(自然科学版)》
CAS
北大核心
2007年第2期182-185,共4页
Journal of Guangxi Normal University:Natural Science Edition
基金
国家自然科学基金资助项目(60672018)
厦门大学"985"二期信息创新平台资助项目(0000-X07204)