基于颜色特征的单主体图像感兴趣区域提取方法  

Method for Extracting Region of Interest in Single Subject Image Based on Color Features

作　　者：邓冠男[1] 贾永鑫 刘乙鑫 Deng Guannan;Jia Yongxin;Liu Yixin(Science College,Northeast Electric Power University,Jilin Jilin 132012)

机构地区：[1]东北电力大学理学院,吉林吉林132012

出　　处：《东北电力大学学报》2020年第4期89-94,共6页Journal of Northeast Electric Power University

基　　金：吉林省自然科学基金(20180101064JC)。

摘　　要：图像检索是指从图像数据库中搜索具有与用户要求相似特性的图像资源.为了提高图像检索的准确率,文中针对单主体图像的感兴趣区域进行研究,首先提出了一种基于颜色梯度的单主体图像感兴趣区域提取方法,能准确的定位并提出感兴趣区域.但是此方法容易造成图像过分割现象,漏掉部分的感兴趣区域的主体.于是在此方法的基础上加以改进,提出了基于区域颜色梯度的单主体图像感兴趣区域提取方法.相较于其他的方法,文中所提出的方法思想简单,能快速有效的实现图像感兴趣区域的提取.Image retrieval refers to searching the image database for image resources with characteristics similar to user requirements.In order to improve the accuracy of image retrieval,this paper studies the region of interest of single-subject image.Firstly,a method of extracting the region of interest of single-subject image based on color gradient is proposed,which can accurately locate and propose the region of interest.However,this method is prone to cause over-segmentation of the image and misses parts of the main body of the region of interest.Based on this method,a method for extracting regions of interest in a single subject image based on region color gradient is proposed.Compared with other methods,the method proposed in this paper is simple and can quickly and effectively extract the region of interest of the image.

关 键 词：图像检索 感兴趣区域 单主体图像 颜色 差分 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]