Citrus Industry July 20186

Artificial intelligence (AI) is increasingly common in electronic devices at home or work, in social media,

video streaming services, electronic commerce, and in internet search engines. Now, AI is rapidly entering the farming scene.

Growers using modern precision agriculture tools and techniques often face a barrage of high data vol-umes created by increasingly prolific, data-hungry electronic devices and services. Compare a smart phone’s data needs with an old desktop phone. Or contrast an old-style paper map of your farm with today’s digital geo-graphic information system maps, showing multiple layers of every square inch of your fields, updated every week or month by automated aerial surveys with drones.

Successful use of precision agri-culture requires improved, smarter methods for managing, understanding and integrating all the “big data” being generated. Fortunately, AI imple-mented as machine learning may be able to help.

Machine learning uses computer algorithms to parse data, learn from it and make determinations without

Artificial intelligence for detecting citrus pests, diseases and disordersBy Arnold Schumann, Laura Waldo, William Holmes, Gary Test and Tim Ebert

human intervention. Since about 2012, new machine vision techniques using deep-learning convolutional neural networks (DL-CNN) have excelled in image recognition, espe-cially in the detection (identification and localization) of objects within images (Figure 1).

DL-CNN models can be trained

to recognize virtually anything that the human eye can perceive, including objects that are partially obscured. In recent global competitions, the accu-racy of DL-CNN machine vision in tests with hundreds of thousands of images has matched or exceeded the accuracy of human viewers. The speed at which machine vision can process

Figure 1. Identification and localization of three objects in an image (pjreddie.com/darknet/yolo)

TM

Mark White: (239) 214-1072William “Buddy” Putnam, IV: (863) 205-1923

Reese Harvin: (863) 781-2997

900 Cowboy Circle, Labelle, FL 33935Phone: 863-675-1500

www.GPSolutionsFL.com Companion, Essential, Nitro, are registered trademarks of Growth Products, Ltd.

Free and fast same-daydelivery

A HEARTY APPETITE!

G.P. Solutions quality and strength is like comparing Rib-eye to a Twinkie, so call today to plan asuccessful menu.

There are few things more nourishing and comforting than a home cooked meal! Citrus trees under intensive stress from the effects of Greening need a HighIntensity Nutritional diet to ensure buds and fruit stay on the trees so ultimately they can produce exceptionally high yields. Yes they can… even with HLB!

Pure and wholesome “TKO” Phosphite 0-29-26 and Recover RX 3-18-18 are based on a potent recipe of technical grade materials that are not just stirredtogether, but bonded with high heat. Sprinkle in Companion® Biological Fungicide as an immune system trigger for a guaranteed disease fighting recipe.

Multi-vitamins such as Zn, Mn and B are showcased in our extensive, chloride and sodium free, MicroTech line of 100% chelated micronutrients for high-end sprays will surely get the physicians seal of approval, keeping critical micros levels boosted.

Roots deserve a “piece of the pie” too; greening can demolish much of the root mass before appearing above ground! Feed them with BioNutrients Solu-ble AG 8-1-9 loaded with organics including humic acid, L-amino acids, kelp and probiotics.

Humic SubstancesFrom Leonardite

Potassium CarbonateNon-Chloride Source

PO3 Phosphite Chemistry

Chelat ion Of Iron

Methylene UreaPolymeric Process

Benef ic ia l BacteriaYeast, Mycorrhizae

GP Solutions Full Page Ad 2017_Layout 1 5/22/2017 4:55 PM Page 1

Citrus Industry July 20188

thousands of images greatly exceeds the capacity of humans.

AG APPLICATIONSSome of the leading uses of

machine vision in precision agriculture are to process video from cameras for identification of weeds, pests, diseases and hazards, and to enhance auto- steering guidance of farm vehicles. Smart sprayers for precision spraying of herbicide only on weeds in agronomic crops have already been developed. They rely on AI machine vision to identify every plant on the ground and make on-the-go decisions about which plants to spray. Herbicide savings of 90 percent were reported from the use of the AI sprayer technology. Development of similar technologies for specialty crops is currently in progress.

Plantix, a free crop advisory app that uses DL-CNN machine vision, was developed in Germany for Android smartphones. At this time, the multi-lingual app is not available yet on the iPhone iOS platform. Plantix includes 26 diagnosed citrus maladies, including citrus canker and greening diseases.

Contact your local Timac Agro USA Sales Representative or a participating retailer. www.us.timacagro.com

BIOSTIMULANT + NUTRITION Improves early tree development

GZ

Use with Generic Imidacloprid: 2% rate, 90 oz in 35 gal Admire Tank.

Cost per tree: $ 0.063.

Dozens of other crops, pests and dis-eases are included in the Plantix library.

There are similar initiatives under-way to develop new and exciting smart agricultural applications for machine learning and AI-enabled computer vision on various mobile computer devices and farm machinery. At the University of Florida Institute of Food and Agricultural Sciences (UF/IFAS) Gulf Coast Research and Education Center, Nathan Boyd is developing diagnostic apps for Florida specialty crops using smartphones as well as AI vision-equipped smart sprayers

for vegetables and strawberries. At the UF/IFAS Citrus Research and Education Center, AI research is focus-ing on autonomous robotic scouting platforms for citrus under protective screen (CUPS) groves. All these proj-ects rely on DL-CNN machine learning for rapid, real-time analysis of video feeds and photographs to identify sit-uations and objects of interest such as weeds, pests and diseases.

ACP DETECTION IN CUPSThe principal reason for growing

citrus in protective screen houses is to

See artificial intelligence in actionThese video clips show artificial intelligence (AI) being used in agriculture:

•• Citrus tree trunk detection with AI: www.youtube.com/watch?v=BRQoP52cFQQ

•• Pest detection with AI: www.youtube.com/watch?v=jBJ1cbZt4KI

•• An agricultural solar-powered robot in action: vimeo.com/140036163

Citrus Industry July 2018 9P536-052384-8 (PCDT Irrigation Southern Print)Color: 4/color Non-BleedSize: T - 4.625 x 10 SIZE D Citrus Industry

80 70 70 10010.2 7.4 7.4 100 100 100100 100 60 100 100 70 70 30 30 100 100 60 100 100 100 10070 70 30 30 100 100 60 70 70 4070 70 30 30 100 40 100 40 40 100 10 40 40 20 70 70 3.1 2.2 2.270 40 40 75 66 6650 40 4025 19 19B 0 0 0 0

100 70 30 100 10 25 50 75 90 100100 60 100 70 30 100 60 40 70 4070 30 100 40 40 100 40 100 40 70 40 70 40 40 340 70 40 70 40 40100 60A

3%ISO 12647-7 Digital Control Strip 2009

PROPANE CAN DO THAT. Don’t let a little thing like bad weather or power outages stop you from farming the way you know best. With a propane irrigation engine you have reliable, effi cient power anywhere and anytime you need it, whether you’re putting water on your crops or pumping it off . What’s more, propane slashes your greenhouse gas emissions and it won’t contaminate soil or groundwater. Take care of business, and the environment? Propane can do that.

PROPANECANDOTHAT.COM

STRONG ENOUGH TO STAND UP TO MOTHER NATURE. CLEAN ENOUGH TO TAKE CARE OF HER, TOO.

exclude the Asian citrus psyllid (ACP), which is the vector of citrus green-ing disease. Therefore, the integrity of the protective screen house must be ensured at all times. Occasional incursions of ACP into the CUPS may be possible, and if not detected imme-diately, they could quickly colonize substantial areas of a CUPS, with sub-sequent spreading of greening disease.

Human scouts are currently the only methods for detecting pests and diseases in citrus groves, and the known methods for ACP detection are very inefficient. Only a small fraction of trees in a grove are sam-pled, and for large trees, only a tiny fraction of each canopy is sampled, with most of the canopy above 6 to 7 feet remaining un-scouted. As a result of these limitations and its laborious nature, scouting is often conducted only weekly or less fre-quently. Consequently, the probability of non-detected psyllids being present during an incursion in CUPS remains high. A more efficient, rapid and com-prehensive robotic scouting method using machine vision and AI is being developed for daily inspection and early detection of ACP and other pests and diseases in CUPS.

A robotic scouting vehicle was designed to be energy self-sufficient, using solar-electric power to move autonomously through the grove with-out human intervention (Figure 2). If you truly want to see and understand

Figure 2. First prototype of a solar-powered robotic scout developed for citrus under protective screen groves

Citrus Industry July 201810

everything in a grove, you should live in it or close to it. The same philosophy should apply to the robotic scout, so it should stay in the grove 24/7, working whenever the sun shines.

Preliminary results of some trained DL-CNN models are shown in Figure 3. Training of new DL-CNN models typically requires 500 to 2,000 digi-tal images that feature the objects of interest for detection. These include arthropod pests such as psyllids, leafminers, mealybugs, scale insects, thrips, spider mites and rust mites, and diseases like citrus canker, greasy spot, anthracnose, melanose and scab.

INDENTIFYING OTHER OBJECTS OF INTEREST

Images of additional identifiable physiological events such as flowering, color break and fruit damage (such as sunburn and splitting) will also be collected as they occur. These other objects of interest are additional iden-tification tasks that a robotic scout could accomplish while moving about to enhance the operation of an inten-sively managed CUPS.

UF/IFAS researchers will also explore other tasks that the scout can perform with machine vision, such as searching for holes in the screen roof and walls, detecting blocked or defec-tive microjet sprinklers, and measuring tree growth. Examples of scouting tasks are shown in Figure 3.

The robotic scout will be equipped with a global positioning system (GPS), Wi-Fi and cellular communi-cations networks, and various cameras of different resolutions. The main computer includes a fast graphics pro-cessing unit for running the DL-CNN object recognition software, while smaller computer peripherals are responsible for solar-power regulation, speed control, steering, navigation and communication. When fully opera-tional, the scout will be guided by both computer vision and GPS waypoints

delineating the intended scouting paths through the grove.

SUMMARYDL-CNN on modern computers

combined with digital cameras makes machine vision that rivals the capabil-ities of human perception to recognize objects of interest. AI machine vision sensors are being combined with autonomous robotic vehicles to improve daily grove scouting in CUPS structures. Although computer AI and machine memory is not even close to the capabilities of human brains, the neural networks are very capable of specific trained tasks that they have been “imprinted” with. A specialized robotic scout could feasibly work faster and for longer hours than human scouts. Future research will compare human and machine scout-ing abilities.

Acknowledgement: This mate-rial is based on work supported by the National Institute of Food and Agriculture, U.S. Department of Agriculture, under award number 2018-70016-27387.

Arnold Schumann (schumaw@ufl.edu), Laura Waldo (ljwaldo@ufl.edu), William Holmes (wdholmes@ufl.edu), Gary Test (gkt@ufl.edu) and Tim Ebert (tebert@ufl.edu) are at the UF/IFAS Citrus Research and Education Center in Lake Alfred.

Figure 3. Machine vision with artificial intelligence can identify important visual attributes of citrus groves linked to pests, diseases, physiological events and disorders.

Detection app optionsThere are three free iPhone apps from the Apple Store for image and object detection. They are PictureThis – Plant Identifier, AlphaPic and What Objects. The first one requires an internet connection (image processing is done on a remote server). The other two apps process everything on the iPhone, which is preferable. In a quick test, the Plant Identifier was able to correctly identify 10 out of 10 houseplants. The Plantix app for Android devices can be downloaded from Google Play.