【摘要】 农作物病虫害症状的检测和鉴定是保证农作物良好生长的前提条件，是人们能够准确、及时地制定防治方案和采取相关措施，切实减轻病虫害的发生。于是提出了以Alexnet为基础的农作物病虫识别方法，首先对采集到的病虫和健康叶片图像进行归档分类，然后对建立好的数据集进行尺寸归一化和数据强化等预处理，最后对训练集采用Alexnet模型进行训练，经过5次训练，实验证明其准确率可达96.93%，该方法能较好地识别农作物病虫害，具有较好的鲁棒性和较高的精确度。更多还原

【Abstract】 Detection and identification of the symptoms of crop diseases and insect pests,so that people can accurately and timely formulate control plans and take measures to effectively reduce the occurrence of diseases and insect pests,which is a prerequisite for ensuring good growth of crops. In this paper,a crop pest identification based on AlexNet is proposed. First,the collected images of pests and diseases and healthy leaves are archived and classified,then the established data set is preprocessed by size normalization and data enhancement,and finally the AlexNet model is used to train the training set. The experiment shows that the accuracy rate can reach 96.93% after 5 times of training,and the method can complete the task of crop pest identification. Convolutional neural network identification technology will become an important way to identify crop diseases and insect pests in the future,which is of great significance to the future development of precision agriculture and modern agriculture.更多还原