Location: Environmental Microbial & Food Safety Laboratory
Project Number: 8042-42000-021-019-S
Project Type: Non-Assistance Cooperative Agreement
Start Date: May 1, 2026
End Date: Apr 30, 2027
Objective:
Accurate and timely detection of citrus diseases is essential for sustaining the productivity and economic viability of the Florida citrus industry. The objective of this agreement is to develop hybrid edge-cloud AI-enabled hardware and software platforms for optical detection and diagnostics of citrus Huanglongbing (HLB), black spot, and other emerging citrus diseases in both field and packinghouse environments.
Approach:
The Citrus Disease Monitoring System (CDMS) will comprise an OBSERVER module deployable across multiple platforms, including smartphone-based handheld devices, ground rovers, and autonomous drones, with connectivity to either EDGE-AI or CLOUD-AI systems for real-time disease diagnostics. The OBSERVER will integrate a multimodal optical sensing suite incorporating RGB, RGB-depth, and multispectral imaging modalities. An enhanced benchtop line-scan hyperspectral imaging system, augmented with RGB-depth capability, will be developed to acquire high-quality image datasets from fruit and leaf samples exhibiting diverse disease conditions. These data will be used to identify optimal spectral wavelengths and features for subsequent multispectral imaging and deployment. The EDGE-AI and CLOUD-AI components will perform data preprocessing, cleaning, and enhancement of multimodal imagery, leveraging advanced sensor fusion and deep learning techniques. These approaches will enable robust classification and accurate diagnostics of in situ fruit and tree canopies in field environments, as well as fruit samples in packinghouse operations.