In order to approve the capability of image retrieval and recognize for color image, an algorithm is proposed that it extracts the local color feature of the central region of image and overall feature, which is based on the accurate modeling of the marginal distribution of wavelet coefficients using generalized Gaussian density ( GGD) . To evaluate the effectiveness of the proposed method, a series of experiments were carried out on two color image database : View and Corel. The experiments show that the method proposed in this paper has more retrieval rate and is more efficient than single algorithm of those.%为了提高广义高斯分布(GGD)在彩色图像检索中的检索性能和图像识别能力,提出了利用局部特征和全局特征的方法,计算图像中心矩形区域的颜色特征作为局部特征和整个图像应用小波变换后应用统计方法,用广义高斯分布计算小波系数的准确分布,提取全局特征,在Corel和VIEW图像库中进行了检索实验.试验表明:该方法的检索能力优于其单一算法.
展开▼