基于颜色特征的茄子病害图像检索方法

为克服当前蔬菜病害诊断专家系统依靠文字提供信息的不准确性，提出一种基于颜色特征的茄子病害图像检索方法。详细分析茄子病害图像的颜色特征，在符合人眼视觉特征的HIS颜色空间上进行图像预处理。提取图像的均值、方差、偏度、峰度、能量作为检索特征值，构建图像特征矩阵。利用高斯归一化方法进行归一化处理后，采用欧式距离进行相似性度量。在Visual C＋ ＋6．0开发环境下，采用C＋＋编程开发了基于颜色特征的茄子病图像检索系统。结果表明，基于颜色特征的茄子病害图像检索方法的查准率为65％，查全率为83％，识别效果较好。将该算法应用于茄子病害诊断专家系统，将大大提高系统的鲁棒性，能够满足病害诊断的要求。

Eggplant Disease Image Search Method Based on Color Characteristics

In order to overcome the inaccuracy in the vegetables disease diagnostician system which relies on text information,a retrieval method was put forward based on color feature for eggplant disease images.The color characteristics of the eggplant diseases image were analyzed and the image pre-processing was proceeded in the HSV color spaces in keeping with human vision.The average value,variance,deviation,kurtosis,and energy were taken as eigenvalues to structure the retrieval matrix.The similarity measurement...