中文说明： 本文提出一种基于HSV空间的颜色直方图特征的图像检索的改进算法，这种改进的方法采用符合人类视觉特性的颜色空间来表示图像的颜色特性，利用柑橘叶片病斑颜色信息不同，借鉴传统直方图方法的检索原理，根据图像颜色非均匀量化后所形成的归一化主色矩阵作为特征进行检索，用户可以通过这种方式来表示查询的视觉要求，与传统直方图方法相比，此方法非常有效，在柑橘叶片病害监测方面较大的提高了检索准确度。
English Description：
In this paper, an improved image retrieval algorithm based on HSV space color histogram features is proposed. This improved method uses the color space that conforms to human visual characteristics to represent the color characteristics of the image. Using the different color information of citrus leaf lesions, and referring to the retrieval principle of traditional histogram method, according to the normalized dominant color matrix formed by image color non-uniform quantization Compared with the traditional histogram method, this method is very effective and greatly improves the retrieval accuracy in citrus leaf disease monitoring.