摘要
在基于内容的图像检索中,需要描述图像中的空间信息从而克服仅基于全局特征的传统方法的局限.图像分割是得到图像空间描述信息的重要手段,但对于所有图像利用同一种分割结果的检索算法(单层描述方法)会受到图像分割算法精确度的影响而使性能受到限制.本文提出了基于图像的两层描述(包括粗略描述和精细描述)和非对称区域匹配的算法以减少不精确分割带来的不利影响.利用从70 0 0张通用图片库中随机选取的70 0幅查询图像而进行的统计实验结果表明此算法可以有效的提高检索效果.
Research on integrating spatial information into content-based image retrieval is aimed at solving the problem caused by global feature based algorithm. Most systems derive the spatial information from image segmentation. However, the description of images based on one-level description (OLD) and the inevitable inaccuracy of segmentation results limit the performance. The proposed two-level description (TLD) describes images by a rough description and a detailed description to avoid improper spatial constrain caused by OLD. Similarity measurement based on unbalanced region matching (URM) is introduced in taking the advantage of TLD to reduce the influence of inaccurate segmentation. The performance of the system is illustrated by experimental results with 700 query images randomly selected from a database of 7000 general-purpose images.
出处
《电子学报》
EI
CAS
CSCD
北大核心
2005年第4期725-729,共5页
Acta Electronica Sinica
基金
国家自然科学基金(No.60172025)
关键词
基于内容的图像检索
两层描述
非对称区域匹配
Algorithms
Content based retrieval
Database systems
Feature extraction
Image segmentation
Measurements