AI-powered remote sensing system for crop monitoring, sizing, and nutrient deficiency assessment

Overview

This solution offers an automated plant phenotyping system that uses drone-based remote sensing and artificial intelligence to assess crop health, growth, and nutrient status. Plant phenotyping, the evaluation of traits such as height, leaf characteristics, canopy width, and yield, is critical for modern agriculture but remains labor-intensive and difficult to scale. By combining aerial imagery capture with advanced image processing, the system enables real-time monitoring of large fields, early detection of nutrient deficiencies, and precise identification of affected plants. The approach supports precision agriculture by allowing targeted fertilizer application, reducing nutrient waste, and improving overall crop yield.

The technology has been validated in rice and cotton fields and is being expanded through an industry partnership to include corn and additional crops. It is designed for agricultural researchers, farm operators, and agronomy companies seeking data-driven tools to optimize crop management.

Technical specifications

Core capabilities:

• Drone-based RGB imagery collection at varying heights to capture field-level and plant-level detail
• Image processing pipelines that segment individual plants from background imagery
• Visual analysis to detect color and pattern changes associated with nutrient deficiencies
• Trait quantification for plant height, canopy width, leaf characteristics, and growth metrics
• Cross-validation of deficiency detection results against laboratory biological analysis

Nutrient coverage (planned and current):

• Nitrogen deficiency detection (validated in cotton and rice)
• Expansion to phosphorus, potassium, magnesium, iron, zinc, and calcium

Data sources:

• A commercial research farm facility and LUMS Nursery in Lahore for ongoing data collection across corn, rice, and cotton crops

Technology readiness level

The system has been experimentally validated in rice and cotton fields, with drone-collected imagery processed to identify nutrient deficiency symptoms and cross-checked against biological laboratory results. Current development is focused on expanding the nutrient deficiency library to include macro and micro nutrients and broadening crop coverage. An active MoU with an industry partner supports continued data collection and tool refinement. The team is seeking collaboration partners to provide equipment such as drones and sensors, as well as funding for key personnel to advance the system toward broader deployment.