Optimizing Agricultural Yields using AI for Automated Spray pump control
- 0 Collaborators
Crop yields are highly affected by the presence of weeds. In small farms, farmers are forced to handpick and uproot weeds as an alternative to use of herbicides. This becomes a challenge differentiating weeds during early stages and the process of handpicking affect the desired crop. In addition, the large farmers are forced to bear with depreciated yields due to presence and effect of weeds. By providing an AI solution to identify the weed at an early stage, we could be able to optimize yields by controlling weed using selective spraying. The main objective of this project is to provide an AI solution that could be used in automated spray pumps to control the weed. The projected solution is a vision detector and classifier model to weed identify and control the weed. This project will lead to a published paper and a product. ...learn more
Project status: Under Development
Internet of Things, Artificial Intelligence
Groups
Student Developers for AI,
DeepLearning,
Movidius™ Neural Compute Group,
Artificial Intelligence Kenya
Intel Technologies
AI DevCloud / Xeon
Overview / Usage
This project aims to develop an optimal solution to solve the weed, crop pests and disease challenge in Agriculture. This solution could be achieved by implementing an AI controlled selective spray pumps which accurately detect and control weeds, crop pests and diseases. The solution will offer a cost-efficient solution to improve crop yields and reduce environment pollution via chemicals which are causing climate change.
Methodology / Approach
The project implements a vision classifier based on leaf features to detect and identify a weed, crop pest or diseases. The research will implement a selective and automated spray pump controlled by the model prediction.
The project is under development and more details will be provided.
Technologies Used
Intel(R) AI DevCloud, Intel(R) Movidus Neural Compute Stick