摘要
提出一种分块局部二值模式的图像检索方法,首先利用3×3像素邻域的局部灰度均值代替其中心像素作为阈值计算LBP值,以改善传统LBP算子的缺陷;然后对图像采用分块处理,以便更好地提取图像的局部特征.实验表明,与传统的基于局部二值模式的图像检索方法相比,此方法具有更高的检索准确率.
In this paper ,a method of image retrieval based on block local binary pattern is proposed .To improve the original LBP ,the average local gray level of a 3 × 3 local area is used to calculate the value of LBP instead of the gray value of its central pixel .Then ,the image is divided into several different sub-blocks .Thus the better local features of the image can be extracted .Compared with the method of original LBP ,the result of experiment shows that the method of image retrieval based on blocked local binary pattern can achieve higher precision .
出处
《微电子学与计算机》
CSCD
北大核心
2014年第5期21-23,27,共4页
Microelectronics & Computer
关键词
图像检索
纹理特征
局部二值模式
分块
image retrieval
texture feature
Local Binary Pattern
block