Agriculture occupies an important position in the economic development, and which is one of the important sources of earning for human beings in many countries. However, a number of challenges are faced by the farmers including different diseases for plants, as having diseases in plants are quite natural. The primary premise in the prevention and treatment of diseases is to correctly identify and judge the disease during the growth of plants, so as to further grasp the rules of disease development. As a matter of fact, an automated plant disease detection technology can be more beneficial for monitoring the plants, and the leaves of plant are the first source of plant disease detecting, the most diseases may be detected from the symptoms appeared on the leaves. Therefore, this paper introduces a novel approach for the automatic detection and classification of plant leaf diseases. Based on the image process techniques, we perform the feature engineering analysis and build the index system for the prediction models. Then the selected features are fed to the GMDH-Logistic model, and the comparison experiments are performed. The outcomes illustrate that the method is effective, it can identify whether the plant is the diseased plant or not.

农业在经济发展中占有重要地位，是许多国家人类重要的收入来源之一。然而，由于植物中的疾病是很自然的，农民面临着许多挑战，包括植物的不同疾病。预防和治疗疾病的首要前提是在植物生长过程中正确识别和判断疾病，从而进一步掌握疾病的发展规律。实际上，一种自动的植物病害检测技术可能更有利于监测植物，而植物的叶子是植物病害检测的第一个来源，大多数疾病可以从叶子上出现的症状中检测出来。因此，本文介绍了一种自动检测和分类植物叶片疾病的新方法。基于图像处理技术，我们进行特征工程分析并为预测模型建立索引系统。然后将选定的特征馈入GMDH-Logistic模型，并进行比较实验。结果表明该方法是有效的，它可以识别植物是否是患病植物。