摘要
提出了一种基于文本、语义和特征块匹配相结合的综合图像检索方法.首先,将图像入库时进行人工标注;然后运用SVM机器学习框架,建立先验知识库,提取图像的语义特征;然后利用Harris算子检测出图像的特征点,进一步统计出两个图像中匹配的特征块数目,计算图像间的相似距离.实验结果表明,这种综合检索方法能更全面、更精确地描述了图像的视觉信息,具有较高的检索效率.
A multi -technique retrieval method was put forward with text -based, semantic and feature. First of all, artificial marks are made when the images are storaged; then SVM machine is used to learning framework＇establishing prior knowledge and extracting the semantic features of the image. Finally, salient points in an image are dected by Harris, the similar distance between two images are calculated based on counting the number of matched salient blocks between two pictures. Experimental results show that the multi -technique can be more comprehensive and precise as describing the visual information of a picture and the method can be highly effective .
出处
《哈尔滨师范大学自然科学学报》
CAS
2011年第1期33-36,共4页
Natural Science Journal of Harbin Normal University
基金
黑龙江省教育厅科学技术研究项目资助(11551435)
关键词
图像检索
语义特征
颜色矩
特征向量
块匹配
Image retrieval
Semantic feature
Color matrix
Attribute vector
Salient blocks matching
