摘要
基于内容的图像检索一直是图像检索领域的研究热点。提出一种综合颜色空间特征和纹理特征的检索算法。首先将彩色图像转换到HSV颜色空间,进行非等间隔量化,提取颜色特征和颜色空间位置信息,再在灰度图像里利用中心块图像分形编码的方法,获取分形编码参数作为图像的纹理特征。在对特征进行内部和外部高斯归一化后,综合颜色空间特征和纹理特征,利用人工鱼群算法进行检索。实验结果表明,算法性能良好,在检索准确率和速率上取得了良好的效果。
Content-based image retrieval is an important and fascinating point in the image retrieval.An image retrieval algorithm based on multiple features was proposed in this paper.First,the color image was transferred into HSV color space,quantized by unequal quantification method,and extracting color feature and each color position message.Then,we implemented image fractal coding based on the center area of the original gray image,and fractal coding parameters are used as texture features.After normalizing each component of the vector and different features by Gaussian normalization method,we combined the color features and the texture features for image retrieval using artificial fish swarm algorithm.The experiment results show that the algorithm can produce good performance in terms of the retrieval effect and the retrieval speed.
出处
《计算机科学》
CSCD
北大核心
2014年第S1期118-121,共4页
Computer Science
基金
中央高校基本科研业务费基金项目(CDJXS12180008)资助
关键词
图像检索
颜色空间信息
分形参数
人工鱼群算法
Image retrieval,Color position,Fractal parameter,Artificial fish swarm algorithm
