基于区域信息的交互式图像检索  

Interactive Image Retrieval Based on Region Information

作　　者：江佳翼 左劼[1] 孙频捷[2] JIANG Jia-yi;ZUO Jie;SUN Pin-jie(College of Computer Science,Sichuan University,Chengdu 610065;Shanghai University of Political Science and Law,Shanghai 200000)

机构地区：[1]四川大学计算机学院,成都610065 [2]上海政法学院,上海200000

出　　处：《现代计算机》2020年第1期56-60,共5页Modern Computer

基　　金：国家重点研发计划项目（No.2017YFC0820103）

摘　　要：基于内容的图像检索(CBIR)是计算机视觉的一个重要方向。由于语义鸿沟的存在,图像检索很难准确捕捉用户需求,交互式图像检索能利用用户反馈在不断交互中改善检索结果。已有的交互式图像检索方法将图像看成整体,忽略了图像中各个区域的重要性。为了更好地捕捉用户对于目标区域的偏好信息,提出一种基于区域信息的交互式图像检索框架,有效地利用图像的区域信息来改善检索性能。实验表明,该方法能较大地提升交互式图像检索的性能。Content-Based Image Retrieval(CBIR)is an important part in commuter vision.Because of the semantic gap,it’s hard to find users’needs accurately.Interactive CBIR can improve retrieval results by using users’feedback.Current interactive CBIR solutions take the image as a whole,ignoring the importance of regions in the image.In order to better capture the user's preference for the target region,Proposes an inter⁃active image retrieval framework based on the region information,which effectively utilizes the region information of the image to improve the retrieval performance.Experiments show that this method can improve the performance of interactive image retrieval.

关 键 词：图像检索 交互式 用户反馈 区域信息 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]