Realizing the Benefits of Precision Agriculture

Realizing the Benefits of Precision Agriculture

UAV with LiDAR camera

This unmanned aerial vehicle (UAV) is equipped with a LiDAR camera.
Photo courtesy of UF/IFAS

Dr. Yiannis Ampatzidis recently joined the faculty at UF/IFAS Southwest Florida Research and Education Center (SWFREC) in Immokalee. He is an Assistant Professor of agricultural and biological engineering. He brings expertise in precision agriculture technology, dedicating part of his time to Extension and the rest on research at SWFREC. I asked a few questions to learn more about Ampatzidis and his work.

What will your UF/IFAS programs focus on?
Ampatzidis: My Extension program in precision agriculture includes development of an educational program to promote adoption and evaluation of state-of-the-art equipment and techniques with a goal of improving the profitability and environmental sustainability of Florida’s fruit and vegetable industry. It integrates knowledge from multiple academic domains, particularly in aerial systems and advanced sensor technologies for citrus and vegetable production.

Yiannis Ampatzidis

Yiannis Ampatzidis

My research program involves unmanned aerial vehicles (UAV) for agriculture and natural systems, smart sensors and machinery, mechatronics, artificial intelligence and robotics, machine vision and learning, automation, remote sensing, wireless sensor networks, and big data applications. Emphasis is being given to developing smart machines and equipment for site-specific applications (e.g., precision sprayer for pests and weeds) in order to reduce agricultural inputs such as water, fertilizer, and pesticides. This research area includes the mechanization and automation of specialty crop production (e.g., harvest), focusing on the design, development, and testing of sensors and control systems for optimal management of inputs, resources, and products.

In general, how has the adoption precision agriculture technology been among citrus growers?
Ampatzidis: We have developed a survey to evaluate the adoption of Precision Agriculture Technologies (PAT) in Florida. This study investigates the factors that affect growers’ adoption of PAT in our state. This analysis could provide a richer understanding of the factors affecting PAT adoption and help determine the importance of weather, risk, and field-specific variables. These factors could then contribute to conservation policy design related to PAT.

Specialty crop growers can take the online survey (10 to 15 minutes) here. NOTE: We will not use/publish personal information.

Do you believe that management of HLB allows more opportunities for growers to adopt precision ag practices? If so, why?
Ampatzidis: Yes, the citrus industry is in critical need of new technologies and practices to manage HLB-infected trees, reduce inputs and production cost, and increase profit. Advances in farm technology (and PAT) can help growers use inputs more efficiently and reduce environmental impacts. They also can help to improve nutrient and water management, reduce competition from weeds, and optimize pest management tactics.

DJI Matrice UAV with Hyperspectral Camera

This DJI Matrice 600 Pro UAV is equipped with a Resonon Hyperspectral camera.
Photo courtesy of UF/IFAS

How can growers utilize imagery of groves, and where can they obtain it?
Ampatzidis: There are a lot of companies developing UAV- or satellite-based solutions for citrus growers. They can fly small UAVs and develop maps that can identify stressed trees, irrigation/water leaks, and create a tree inventory. My Precision Agriculture Engineering program is developing a UAV-based software for citrus growers utilizing artificial intelligence and various sensors such as multispectral and hyperspectral cameras, LiDAR, among others.
Our DJI Matrice 600 Pro is equipped with a Resonon Hyperspectral Camera and a Velodyne LiDAR system to capture images using UAVs.

What is artificial intelligence and how can it be applied to citrus?
Ampatzidis: Artificial intelligence and machine learning are two hot areas in computer science and robotics. Machine learning is an application of artificial intelligence based on the idea that a machine (e.g., computer) can learn from data and identify patterns (in data). That process can eliminate human intervention and errors. In general, a computer can “learn” from data, without being programmed, and adjust to new inputs to accomplish specific tasks (e.g., self-driving cars).
Machine learning and artificial intelligence can change modern agriculture. One example is the smart (precision) sprayer developed by BlueRiver Technology. The smart sprayer utilizes a vision-based system and artificial intelligence to detect and identify single crop plants and weeds and spray only on the weeds. It can save more than 90% of herbicides (comparing with the traditional sprayers).

Another example is the robotic harvester developed by Harvest CROO Robotics; a company located in Tampa. The strawberry harvester detects and locates ripe berries using machine vision and artificial intelligence.

How is disease detection software developed?
Ampatzidis: There are several ways to develop a disease detection system. For example, we have developed a vision-based system utilizing a RGB camera (e.g., cell phone camera) and artificial intelligence to detect symptoms of a disease and discriminate from other disorders or pathogens, despite the strong similarity. This system enables growers to take photos of a possibly affected plant with their mobile device, upload the image from the mobile device or computer, have the image processed remotely through the cloud by a deep learning system, and receive a prompt diagnosis of the specimen. It utilizes a novel algorithm for fusing data at different levels of abstraction to improve performance of the system. The algorithm discovers low-level features from raw data to automatically detect veins and colors that lead to symptomatic leaves. This work shows potential for massive screening of plants with reduced diagnosis time and cost.
Another nondestructive method to a detect disease, even before visible symptoms appeared (e.g., laurel wilt infected avocado trees), and distinguish it from healthy and other diseases or factors that produce similar external symptoms (e.g., nutrient deficiencies) is by using spectral data (e.g., multispectral or hyperspectral sensors) and artificial intelligence.

There has been a lot of emphasis on citrus root health, are there any precision ag applications that could help in this area?
Ampatzidis: We are developing a system (non-destructive method) to map tree roots utilizing a ground penetrating radar and a novel software (funded by the UF Citrus Initiative). We use this system to evaluate HLB-infected citrus rootstocks and to compare rootstock propagation methods.

Meet Meister Media’s GPI
We at Meister Media recently conducted an extensive online survey across multiple agriculture occupations around the world. It showed precision agriculture is catching a wave.

One respondent, a specialty crop grower from Florida, noted simply that precision is “going to continue to gain speed because it is driven by new technologies.” And indeed it is. He and many other growers said they’re particularly excited by developments in precision irrigation, sensors, as well as field and crop imagery.

Precision Ag Specialty Crops logoWe’re of course pleased to hear this. By now, we hope you’ve noticed the increasing occurrence of our “PrecisionAg Specialty Crops” logo in stories such as this one. This is part of a wider effort across all Meister Media titles called the Global Precision Initiative (GPI). We believe precision is going to continue to transform agriculture.

We believe, furthermore, that we all still have a lot to learn, especially from one other. This is the aim of the GPI. We hope you’ll join us for the ride.