基于形状特征的图像检索技术  

Technique of image searching based on shape features

作　　者：刘威[1] 孔英蕾[1] 肖宏涛[1] 

机构地区：[1]河北廊坊广播电视大学开放教育部,河北廊坊065000

出　　处：《兰州理工大学学报》2011年第1期100-105,共6页Journal of Lanzhou University of Technology

摘　　要：提出一种新的基于图像形状的检索方法.首先,用Canny算子对图像进行平滑处理,提取图像边界方向直方图特征;其次,用改进的不变矩来描述图像形状的区域特征,改进后的不变矩特征不受图像的缩放、平移和旋转影响;最后,为了克服不变矩只针对对象区域而忽视图像边界的缺点,提出改进的不变矩与边界方向特征相结合的方法,使得检索取得更好的效果.仿真实验表明,提出的检索方法能够有效提高图像检索的效率.A new kind of image retrieval method based on shape content was proposed.First,a Canny operator was used to smooth the source image and extract the feature of the edge direction histogram.Then,the improved invariant moments were used to describe the shape regional feature.The rotation,scaling-down or-up,and translation of images had no influence on those improved invariant moments.Finally,in order to overcome the shortcoming that invariant moments focus only on the region of image while overlook areas the shape edge characteristics,a new method was proposed where the improved invariant moments and edge direction were integrated.It was shown by the simulation test that the efficiency of image searching could effectively be improved by the searching method presented.

关 键 词：图像检索 CANNY算子 不变矩 方向特征 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]