Hyperspectral camera for plant analysis: agri-photovoltaic and agricultural machine vision AI

• Main Author: Ong, Richard Zhao Ting
• Other Authors: Ng Yin Kwee
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Agricultural Sciences; Engineering
• Online Access: https://hdl.handle.net/10356/178137
• Institution: Nanyang Technological University

Global food security faces challenges from climate change, supply disruptions, and geopolitical conflicts, causing shortages and hunger. Singapore, heavily reliant on food imports, aims to produce 30% of its nutritional needs domestically by 2030, using minimal land, focusing on highly productive and climate-resilient agriculture. To address productivity threats, early detection of pests and diseases is crucial, with modern technology like cameras and sensors streamlining monitoring processes. Hyperspectral imaging, surpassing traditional methods, offers comprehensive plant assessment beyond visible spectrum, potentially revolutionizing plant health monitoring with insights beyond traditional factors. This project will aim to develop a low-cost hyperspectral camera system and integrate machine learning algorithms for image processing. The resulting system is a solution to early detection of pest/disease pressure, growth rate, leaf area, etc. Therefore, corrective action can be taken early, allowing yields to recover from initial underperformance.