摘要
随着大容量图像数据库的广泛应用,基于内容的图像检索(CBIR)技术成为对其进行管理和检索的有效手段。文中提出了一种基于灰度变化进行基于内容的图像检索的方法,利用差分概念提取出图像的差分矩阵,再从中提取特征进行度量。给出了对自然图像库检索的实验结果,实验表明,采用该方法比空间颜色直方图与共生矩阵等方法能取得更好的检索效果。与颜色空间直方图和灰度共生矩阵的比较实验证明差分矩阵的方法在速度和结果上都有比较好的效果。
With the widely usage of large image database, content-based image retrieval(CBIR) becomes an effective method to carry out its management and retrieval. Here puts forward a method of CBIR based on gray difference. This method obtains the difference matrix from the image and then extracts feature vector from the difference matrix. The result of the experiments to nature image database shows that the method of different matrix is more effective than that of spatial color histogram and co-occurrence matrix.
出处
《微机发展》
2003年第11期120-122,共3页
Microcomputer Development
