基于目标区域的彩色图像检索研究

提出了一种基于目标区域的彩色图像检索算法，首先，获取彩色图像HSV色彩空间中各个分量灰度图像，然后计算各分量灰度图像的二值边缘图像，根据边缘图像的连通性提取彩色图像的目标区域，在图像检索时，抽取彩色图像目标区域的提取和基于目标区域的彩色图像检索，实验表明，基于目标区域的彩色图像检索算法优于基于全局图像的彩色图像检索算法。

COLOR IMAGE RETRIEVAL BASED ON OBJECT REGIONS

When an image is observed, it is natural that the image is to be divided into the object regions and the background regions according to the human vision. Generally, the image retrieval performance is evaluated mainly with the object-based measurement. A color image retrieval algorithm based on object regions is presented in this paper. First, each component grayscale image in the HSV color space of the color image is obtained, and then the binary edge image of each grayscale image is calculated. In the color image retrieval based on object regions, it is most important to extract the object regions from the color image effectively. According to the connectivity of the edge images, the object regions are extracted from the color image. During the image retrieval, the features of the subimages, which the object regions of the color image correspond to, are extracted and used in the image similarity matching process in place of the global image features. And two classes of experiments are conducted: the object regions extraction and the color image retrieval based on object regions. Experiments show that the color image retrieval algorithm based on object regions is superior to that based on the global image.