文章摘要
赵珊,郑清洁.基于字符串匹配技术的图像检索算法[J].高技术通讯(中文),2010,20(2):117~121
基于字符串匹配技术的图像检索算法
An image retrieval algorithm based on character string matching
  
DOI:
中文关键词: 基于内容的图像检索(CBIR),方块编码(BTC),字符串匹配,视觉特性
英文关键词: content based image retrieval (CBIR), block truncation code (BTC), character string matching, visual feature
基金项目:国家自然科学基金(50804013),南京邮电大学江苏省图像处理与图像通信重点实验室开放基金(ZK208002)和河南省教育厅科学基础研究基金(2008B520015,2009B520013)资助项目
作者单位
赵珊 河南理工大学计算机科学与技术学院
南京邮电大学江苏省图像处理与图像通信重点实验室 
郑清洁 河南理工大学计算机科学与技术学院 
摘要点击次数: 2659
全文下载次数: 2120
中文摘要:
      为提高图像检索的效果,提出了一种基于字符串匹配技术的图像检索算法。该算法根据人眼的视觉特性及方块编码(BTC)的原理首先对图像进行分割,构造对表征图像内容有意义的图像特征。在此基础上,根据字符出现的概率对字符表征意义的重要性,把图像特征动态映射成字符串形式,然后采用字符串匹配技术进行图像检索。该算法不仅利用了图像中的边缘及纹理分布,而且将字符串匹配技术引入到图像检索中,在提高检索率的同时又加快了检索速度。实验结果表明,该算法具有较高的检索效率。
英文摘要:
      A novel image retrieval algorithm based on the character string matching technology is presented in this paper in order to improve the effect of content based image retrieval (CBIR). The algorithm segments the image first according to the human visual feature and the principle of the block truncation code (BTC), and then extracts the image features important to characterize image contents. Based on this, the image features are mapped into the character string. Then, the technology of character string matching is adopted for image retrieval. The algorithm takes not only the edge and texture information but also the character string matching into account, resulting in the increase in the rate of retrieval. The experimental results show that the proposed method has the sound and robust retrieval performance.
查看全文   查看/发表评论  下载PDF阅读器
关闭

分享按钮