节点文献
科技资源图像检索技术研究与实现
Research and Implementation of Image Retrieval Techniques from Science and Technology Resources
【摘要】 科技资源具有海量、复杂、异构、地理分布广的特点。为了高效查找和使用科技资源的图像,综合运用TB IR(基于文本的图像检索)和CB IR(基于内容的图像检索)共同检索科技资源图像,设计了科技资源图像检索系统,并对系统中的图像分割、特征提取和相似度度量3种关键技术进行了研究。
【Abstract】 In order to search and utilize images efficiently from Science and Technology Resources,which are usually massive,complex,hybrid and distributed.This paper points out how to combine TBIR(text-based image retrieval) with CBIR(content-based image retrieval) to search images from science and technology resources.An image retrieval system architecture from science and technology resources is designed.Three key techniques in this system,which is image retrieval,image segmentation,feature extraction,similarity measurement,are studied.
【关键词】 科技资源;
图像检索;
图像分割;
特征提取;
相似度度量;
【Key words】 science and technology resources; image retrieval; image segmentation; feature extraction; similarity measurement;
【Key words】 science and technology resources; image retrieval; image segmentation; feature extraction; similarity measurement;
【基金】 河北省自然科学基金资助项目(06649046D)
- 【文献出处】 广西师范大学学报(自然科学版) ,Journal of Guangxi Normal University(Natural Science Edition) , 编辑部邮箱 ,2007年04期
- 【分类号】TP391.3
- 【下载频次】34