Agriculture is an important pillar of the national economy, and crop pest control has been an important issue restricting agricultural development. Traditional pest identification and prevention and control methods mainly rely on farmers' experience, which has the defects of low accuracy and low efficiency. With the development of artificial intelligence technology, the use of image recognition and deep learning technology to develop a system AI to identify pests and diseases, can quickly and accurately identify and early warning of crop pests and diseases, improve the efficiency of pest control, has great practical application value. Farmland Watchman-Based on EasyDL's intelligent agricultural identification system with big data analysis as the background, the intelligent agricultural service system is constructed. The system is mainly composed of farmers, experts, administrators of three parts, based on EasyDL platform and Android platform, based on Java language, based on SpringBoot and Postman and other tools to develop system architecture. In view of most of the diseases and insect pests and the use of drugs that cannot be judged in time, our group makes timely adjustments to meet the needs of farmers to the greatest extent, solve the problem that farmers cannot spare a lot of time for systematic learning, diagnose diseases and insect pests in a timely and accurate manner, expand the application field of intelligent agriculture in the digital age, and fill the gap in intelligent identification in the digital age. This work has four innovations and five advantages: Four major innovations (1) accurate disease diagnosis. Farmers only need to take photos and upload diseased crops through mobile phones, and according to big data analysis, accurate results can be returned to farmers in 1-2 seconds. Currently, more than 70 kinds of common diseases and insect pests can be identified, covering more than 30 kinds of common crops, and the identification accuracy can reach 94.64. (2) the way of learning is convenient. The system provides an online learning platform. There are rich learning resources on the platform, including pest identification skills, control methods, pesticide selection, etc., and farmers can learn according to their own needs. (3) Accurate recommended medication. Through the accurate identification of crop diseases and insect pests, the system will recommend the use of drugs and dosage. By reducing the use of pesticides, it can reduce the damage to beneficial organisms and the destruction of the ecological environment, and maintain the ecological balance. (4) the establishment of farmers forum. Farmers can obtain the latest research results and technical skills of pest control on the forum to realize the sharing and dissemination of knowledge. five major advantages (1) to build a database of pests and diseases. This data set will provide valuable basic data resources for the field of agricultural disease recognition research, and can be used as a standard library of ML modeling in the big data environment, which has important practical application value to promote the development of agricultural disease image recognition research (2) independent labeling of details. Through a large number of material data and accurate labeling, the existing model can achieve an accuracy rate of 94.64, greatly reducing the problem of misjudgment by farmers and causing untimely crop control (3) APP construction has been completed (4) agricultural experimental fields for diseases and insect pests on both sides of Alishan. Our team has an agricultural test field for pests and diseases on both sides of Alishan, with a total area of 5213 square meters. This experimental field provides us with valuable practical opportunities, allowing us to combine theory and practice, and continuously optimize our intelligent recognition technology (5) to develop agricultural data visualization platform. Simultaneous development of the Internet of Things big data monitoring platform, through soil sensors, wind speed sensors, moisture sensors and other Internet of Things monitoring equipment to build an agricultural IOT platform, farmers can be based on the data platform to give the environmental analysis, combined with the actual situation of pests and diseases, better from the root causes of pest control. With the continuous development of artificial intelligence technology, more and more fields begin to apply AI technology to improve the efficiency and quality of their respective industries. In the field of agriculture, AI intelligent identification of pests and diseases has become an important application direction. This technology can help farmers find and solve the problems of diseases and insect pests in time, and improve the yield and quality of crops, which has important practical significance and application value.