PROJECT TITLE: Artificial intelligence apps for smartphones: a modern diagnostic extension tool for citrus growers and home owners to rapidly identify nutrient deficiencies and HLB symptoms in Florida groves

OBJECTIVES:

1. Develop fast and accurate diagnostic artificial intelligence models for key nutrient deficiencies of citrus encountered when trees are impacted by HLB disease
2. Transfer the best detection models developed to smartphone platforms (iOS and Android) and develop apps to use the models for a) identifying the above disorders in-field and b) to provide recommendations for their management with enhanced nutrition
3. Demonstrate and distribute the prototype smartphone app to citrus extension agents and citrus growers, and provide initial support by creating extension and outreach materials on UF/IFAS EDIS libraries and by offering a workshop for using the new smartphone app.

Click to open the Citrus Diagnosis App home page

Leaf symptoms already trained for detection by the deep learning artificial neural networks (see demo video below):

Nutrient deficiencies: nitrogen, iron, manganese, zinc, magnesium

Pests: spider mite damage, psyllid damage, thrips damage, leafminer damage

Diseases: HLB, citrus canker, greasy spot, citrus scab, phytophthora

Other: Leaf sunburn / salt damage

Additional leaf symptoms to be trained:

Deficiencies: potassium

Pests: rust mites

Diseases: melanose

Funding: This project was made possible by funding from the HLB Multi-Agency Coordination (MAC) System

Personnel:

Arnold Schumann - professor (schumaw@ufl.edu)

Chris Oswalt - extension agent IV

Laura Waldo - biological scientist III

Perseveranca Mungofa - graduate student