结合对象语义的图像检索方法
New CBIR Method Integrated With Object Semantics
本着缩短图像高层语义与低层特征之间差距，本文提供了一种新的结合对象语义的图像检索方法。文章以应用对象描述模型（ODM）为基础，结合机器转换模型，介绍了通过训练学习得到图像高层语义与低层特征之间关系的方案。本文所提出的方法反映了人的高层语意与图像低层特征之间的联系，达到了通过高层语义对图像库进行检索的目的。
In order to reduce the distance between image high-level semantics and low-level feature, we introduce a new CBIR method integrated with object semantics. Based on the object description model and combined with the machine transform model, we provide a strategy to obtain the relationship between image semantics and features by learning . Experiment shows that the results of our method are much better than the traditional method.
贾振超、赵耀
电子技术应用
对象识别 机器学习 图像检索
Object recognition，machine learning CBIR
贾振超,赵耀.结合对象语义的图像检索方法[EB/OL].(2005-10-25)[2025-08-02].http://www.paper.edu.cn/releasepaper/content/200510-263.点此复制
评论