基于卷积神经网络的茶叶病虫害检测系统设计

Design of tea pests detection system based on convolution neural network

为解决茶农在识别茶叶病虫害时存在的主观性强、误判率高等问题,基于卷积神经网络构建的茶叶病虫害识别模型经过训练、调优后获得了最终的检测模型,该检测模型通过Java Web技术构建成B/S模式的病虫害在线检测系统。用户通过在浏览器中提交待识别的茶叶图像至服务器,服务器将接收到的病虫害图片送入检测模型进行病虫害识别,将识别结果返回至用户端。实验结果表明,基于卷积神经网络构建的茶叶病虫害检测系统实现了茶叶图像中18种病虫害的检测,能较好地帮助茶农快速识别茶叶病虫害,对茶叶病虫害防治具有重要意义。

In order to solve the problems of strong subjectivity and high misjudgment rate of tea farmers in identifying tea pests and diseases,the tea pests and diseases identification model based on convolution neural network was trained and optimized to obtain the final detection model.The detection model was built into a B/S mode online detection system of pests and diseases through Java Web technology.The user submits the tea image to be recognized in the browser to the server,and the server sends the received image of pests and diseases to the detection model for pests and diseases identification,and returns the identification results to the user.The experimental results show that the detection system of tea pests and diseases based on convolution neural network has realized the detection of 18 kinds of pests and diseases in tea images,and can help tea farmers quickly identify tea pests and diseases,which is of great significance for the prevention and control of tea pests and diseases.