为了根据作物不同病害程度等级采取不同防治方法，实现作物高产和减少环境污染，提出了一种复杂背景

下的作物叶片病害等级分类算法。首先，利用阈值分割法对黄瓜病害叶片图像进行病斑分割 其次，计算病斑区域中

像素个数与病叶区域中像素个数的比值 最后用作物病害等级分级标准进行比较来确定病害等级类别。利用该方法

在2 种作物5 种常见病害叶片图像数据库上进行了病害等级分类试验，识别精度高达92． 7 。结果表明，该方法对

作物病害叶片等级分类是有效可行的-To take a different control methods based on different levels of crop disease levels, crop yields and reduce environmental pollution, the proposed plant leaves disease classification algorithm under complex background. First, the threshold segmentation method cucumber disease leaf images lesion segmentation　secondly, to calculate the ratio of the number of pixels in the lesion area and the number of pixels in the region of diseased leaves　and finally a comparative rating of crop disease grading criteria for determining disease level category. Using this method on two kinds of crop disease leaf images using five common　of disease classification test, recognition accuracy up to 92. 7 . The results show that the method of crop disease leaf classification is feasible and effective