A farmland pest and disease identification and early warning method, device, equipment and storage medium
By deploying camera equipment and pest classification models in farmland, combined with temperature and humidity sensors and LSTM networks, efficient and accurate pest detection and early warning were achieved, solving the problems of low timeliness and accuracy in existing technologies.
CN122454292APending Publication Date: 2026-07-24ENTROPY CLOUD BRAIN MACHINE (HANGZHOU) TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ENTROPY CLOUD BRAIN MACHINE (HANGZHOU) TECHNOLOGY CO LTD
- Filing Date
- 2026-05-27
- Publication Date
- 2026-07-24
AI Technical Summary
Technical Problem
Existing methods for monitoring agricultural pests and diseases are not timely or accurate, and are prone to missing the optimal control period.
Method used
Images are acquired by cameras deployed in farmland, and detection is performed using a pre-trained pest and disease classification model. The insect population density and lesion area are counted, and the shooting frequency is adjusted by temperature and humidity sensors. Pest prediction is performed using an LSTM temporal neural network, and an early warning is issued when the pest and disease level exceeds a preset threshold.
Benefits of technology
It improves the timeliness and accuracy of monitoring farmland pests and diseases, enabling timely early warning and reducing farmland losses.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN122454292A_ABST
Abstract
The application discloses a kind of identification and early warning method, device and equipment of farmland pest and plant disease, and storage medium, method includes: obtaining the shooting image of target farmland, input shooting image into pest and plant disease classification model, output detection frame indicating pest and plant disease in shooting image, the total number of all mouthparts in shooting image is counted and the mouth density is calculated, the lesion area of all diseases in shooting image is counted, according to the mouth density in the minimum circumscribed region, and lesion area, determine pest and plant disease grade, and when pest and plant disease grade exceeds preset pest and plant disease grade, carry out pest and plant disease warning corresponding to pest and plant disease grade.
Need to check novelty before this filing date? Find Prior Art