基于灰度和边界方向直方图的医学图像检索

本文研究了采用分级检索的机制,综合利用灰度及形状特征进行基于内容的医学图像检索的方法,该方法克服了灰度直方图不能充分表示空间分布信息的不足。利用边界方向直方图描述形状特征,避开了对图像进行精确分割这一医学图像处理中的难点问题。对CT图像数据库进行的检索实验,验证了该方法具有良好的检索性能。

Medical Image Retrieval Based on Histograms of Gray and Edge Directions

Based on hierarchical indexing model, a new method of content-based medical image retrieval was researched, which is based on both gray and shape information. The limitation of the gray level histogram algorithm, which can't represent the spatial information of the image adequately, was overcome by using this method. The shape feature is described by edge directions histogram. This method avoids image segmentation which is an open problem of medical image processing. A database of CT images was retrieved by this method. The result shows that this method has good performance.