基于嵌入式零树小波编码直方图图像检索

Image indexing Based on Embedded Zerotree Wavelet Coding Histogram

图像和视频应用的快速增长,使得根据图像和视频内容进行查询的技术变得越来越重要。人们提出了许多基于像素域或压缩域的图像检索技术。因为多媒体数据库通常具有相当大的数据量,所以基于像素域图像检索技术的计算复杂度相当大。因此,许多文献提出更快的基于压缩域的图像检索技术。本文提出一种改进的基于嵌入式零树小波编码直方图的图像检索技术。特征提取综合考虑图像的颜色、纹理、频率和空间信息,所有的特征可以在压缩过程中自动得到。图像检索的过程就是匹配待检索图像和来自数据库的侯选图像的索引。实验证明这种方法具有好的检索性能。

Image retrieval plays important role in the image and video applications in recent years. It is crucial to develop indexing technique for searching video and images based on their content. Several indexing techniques have been proposed in pixel and compressed database, but they are hard to be popularized because of their great computational complexity of pixel domain indexing. Compressed domain indexing techniques that provide faster retrieval speed are becoming popular. In this paper, an improved image indexing technique based on embedded zerotree wavelet coding histogram is proposed. The proposed indexing features contain color, texture, frequency, and spatial information of the query image given. All features can be extracted as a byproduct during the image compression. Image retrieval is conducted by matching the index of the query and candidate images from the database. Experiment indicates a good retrieval performance of the proposed technique.