摘要
基于内容的图像检索技术主要是通过提取图像的颜色、形状、纹理等可视特征作为图像检索依据,对图像库进行近似检索.针对海洋生物图像的特点,将离散余弦变换技术应用到图像检索中,计算基于圆环分割的图像块的DCT能量矩得到图像的纹理信息.在此基础上,提出了一种综合DCT能量矩和不变矩的多特征融合的图像检索方法,并将其应用到海洋生物图像检索中.实验结果表明,使用该方法可以提高海洋生物图像检索的准确率,且对图像的旋转、缩放和平移具有不变性.
The image retrieval technology based on content is mainly by extracting image color,shape,texture and other visual features as an image retrieval basis to obtain approximate results from image database.In this paper,according to the characteristics of the marine organism image,the technique of discrete cosine transform is used in image retrieval;the annular energy moment based on DCT is presented as an efficient textural feature descriptor.Based on this,we propose a comprehensive image retrieval method based on DCT energy moment and Hu moment invariant,and apply it to marine organism image retrieval.The experimental results show that the accuracy and speed of the retrieval results are improved,and at the same time the proposed method is very robust in terms of scale and rotation invariance.
出处
《河南大学学报（自然科学版）》
CAS
2015年第2期217-222,共6页
Journal of Henan University:Natural Science
基金
福建省自然科学基金(JK2012026)
关键词
图像检索
离散余弦变换
能量矩
不变矩
image retrieval
discrete cosine transform
energy moment
moment invariant
