sUAS in use at a Tennessee-based nursery.
Photos courtesy of James Robbins

Nurseries are beginning to purchase small unmanned aircraft systems (sUAS), or drones, to accomplish several tasks including sales and marketing, crop inventory, and crop monitoring. Nursery operators are faced with a number of business decisions before adopting this emerging technology. To generate useful images, a nursery would need to purchase an appropriate aircraft, sensor, computer, and image processing software.

Like any other business decision, the nursery needs to evaluate what option makes the most sense based on the cost outlay of each option and the type and frequency of needs that the business has. For example, Nursery A only needs a high-quality video/images of their overall operation which can be used on the company webpage and for use at trade shows. Nursery B wants to use the sUAS for weekly crop monitoring. Nursery A may find that it is more cost effective to pay a third party provider that specializes in flying the sUAS and has the hardware and software to generate the video product. However, Nursery B may determine that it is more cost effective to purchase the aircraft, sensor, and software so they have the capability in-house and can easily fly on-demand.

There are two broad categories of sUAS, rotary (left) and fixed wing. In general, the best aircraft option for nurseries would be a rotary aircraft. Fixed-wing platforms can cover more surface area in the same amount of time, but they are more expensive, less stable for collecting imagery.

Aircraft

There are two broad categories of sUAS, rotary and fixed wing. In a majority of cases, the best aircraft option for nurseries would be a rotary aircraft. While fixed-wing platforms can cover more surface area in the same amount of time (e.g. likely 6 times more) compared to rotary aircraft, fixed wing are more expensive, less stable for collecting imagery compared to rotary, and require some kind of assistance in take-off and landing. Rotary aircraft range from two to eight (octocopter) blades with 4 (quadcopter) being the most common. The advantages of more blades are stability and redundancy. Most sUAS are made of lightweight materials such as plastic, aluminum, and carbon fiber. Power is most often provided by lithium ion Polymer(LiPO) batteries. As a beginner, a ‘bundle’ might be a good purchase decision, so you get spare parts and a carrying case in the package. The most dramatic enhancements in the past three years has been in the flight navigation software. Even very low cost aircraft come with powerful automated flight software to help plan and execute flight plans. On some units, the flight plan can be saved which is especially useful if the flight pattern needs to be repeated. To get a useful professional aircraft, you will likely spend between $800 and $5,000. Some of the more common manufacturers are DJI, Autel, Parrot, and Yuneec. To help one better understand the process and terminology in purchasing an aircraft, a must read is the University of Arkansas fact sheet on ‘Features to Consider When Purchasing a Small Unmanned Aircraft System (sUAS)’. It is worth noting that the ‘s’ in sUAS describes a new category (small) created by the FAA as a part of the permanent guidelines called Part 107. To be consider in this ‘small’ category the entire system (aircraft, sensor, battery) must weigh less than 55 pounds.

Sensor

Understanding that payload weight is critical when flying sUAS, sensor manufacturers have responded by quickly miniaturizing sensors. There are seven categories of sensors (Table 1). A major reason nurseries will limit themselves to certain sensors will be cost. An alternative to owning your own sensor is to hire an outside service for more expensive operations.

Nurseries are most likely to own an RGB camera, modified RGB camera, or multispectral sensor. An RGB camera would be useful to take still images or high-quality video that can be used in sales & marketing, plant inventory, and simple crop monitoring. These low-cost cameras can also be used to safely inspect greenhouse/hoop houses coverings or structural elements. A modified RGB camera would be useful to generate a ‘vegetative crop index’ such as NDVI (Normalized Difference Vegetation Index). To generate a NDVI image requires a sensor with Red and NIR bands and software to process the image. Moving up in cost would be a multispectral sensor which is useful for more complex crop monitoring.

Software

The final piece in your aerial imaging system would be software. Some examples of software that you may use for processing aerial images include open-source or free software such as QGIS, Microsoft ICE, ImageJ, MeshLab and commercial software such as Agisoft Photoscan and Pix4D. Remember, instead of owning the software yourself, another option is to pay a third party provider to process your images. Examples of data processing services include DroneDeploy, Airinov, and Agremo (formerly AgriSens).

Example of the processed images using ICE. Flight 1 (Left) was mostly container-grown plants in the open while Flight 2 (Right) was similar plants under metal hoops (poly sheet removed). See corresponding table above photos.

It is likely that a nursery would want to ‘stitch’ multiple aerial images together into a single composite image. To give you an example of what is involved we recently collected some images at Willoway Nurseries (see photos on this page) and processed them using the free software from Microsoft called ICE (Image Composite Editor).

It is likely that within five years almost every nursery will own a sUAS for even simple tasks such as infrastructure inspection or sales and marketing. Like any capital investment, nurseries need to crunch some numbers to evaluate how deep to extend into this emerging technology themselves or whether it makes better financial sense to hire a third-party provider for specific services.

James Robbins is a professor at the University of Arkansas System CES, jrobbins@uaex.edu. Joe Mari Maja is a sensor engineer at Clemson REC, jmaja@clemson.edu.

No endorsement is implied or discrimination intended for firms or references included or excluded in this article.