Drone Applications for Puerto Rican Agriculture - UNIVERSITY OF PUERTO RICO AT MAYAGUEZ

DRONE APPLICATIONS FOR PUERTO RICAN AGRICULTURE

Sponsoring Institution

National Institute of Food and Agriculture

Project Status

COMPLETE

Funding Source

HATCH

Reporting Frequency

Annual

Accession No.

1009292

Grant No.

(N/A)

Cumulative Award Amt.

(N/A)

Proposal No.

(N/A)

Multistate No.

(N/A)

Project Start Date

Mar 14, 2016

Project End Date

Feb 25, 2019

Grant Year

(N/A)

Program Code

[(N/A)]- (N/A)

Recipient Organization
UNIVERSITY OF PUERTO RICO AT MAYAGUEZ
P. O. BOX 9000
MAYAGUEZ,PR 00681

Performing Department
Agri Engineering

Non Technical Summary
Coffee, plantain and forage grasses are vital crops of Puerto Rico, yet current crop sensing methods fail to capitalize on technological advances that could improve their economics. An unmanned aerial vehicle (UAV or drone) equipped with sensors can be used to address coffee characterization, assessment of crop damage by storms, and forage yield sensing. A LIDAR (light detection and ranging) sensor or multi-spectral camera will be mounted on a drone to scan or image the crops. The collected sensor data will be analyzed and correlated with the field experimental data to address the issues mentioned previously. Results from this work will impact Puerto Rican agriculture in 3 areas: it will enhance the Puerto Rican coffee industry by registering geographical indicators, offer an alternative method for assessing storm and hurricane damage, and allow for planning and sustainable production of forage grasses.

Animal Health Component

75%

Research Effort Categories

Basic

Applied

75%

Developmental

25%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
402	1099	2020	20%
402	2232	2020	15%
402	1649	2020	15%
404	1099	2020	20%
404	2232	2020	15%
404	1649	2020	15%

Knowledge Area
404 - Instrumentation and Control Systems; 402 - Engineering Systems and Equipment;

Subject Of Investigation
2232 - Coffee; 1099 - Tropical/subtropical fruit, general/other; 1649 - Forage legumes, general/other;

Field Of Science
2020 - Engineering;

Keywords

coffee characterization

Goals / Objectives
The goal of this project is to develop capabilities for the use of drones in Puerto Rican agriculture by procuring drones and sensors, and demonstrating their applications. Three possible applications have been identified according to need, availability of resources, and future funding opportunities.Three specific objectives of the proposal are to: 1. develop methods for coffee characterization using drones,2. improve storm damage assessment methods using UAV technology,3. sense forage yield through sensors mounted on a drone.

Project Methods
Following sections briefly describe the methodology to meet three objectives proposed in this proposal.Objective 1: Coffee characterizationImage segmentation methods developed by Zahawi et al. (2015) and Torres-Sanchez (2015) were identified for coffee characterization. Prior to acquiring sensor data, coffee plant phenotypes: age of plantation, variety, plant height, canopy diameter, above ground biomass, chlorophyll content and others will be collected at different crop growth stages. Selected segmentation methods would be applied to coffee images. If needed, an image segmentation method would be developed to meet crop and application specific needs. The GPS locations of the fields will be registered to the acquired and processed images to meet regulatory and promotional needs of Puerto Rican coffee producers. Coffee plant phenotypes, field recorded and determined from sensed data, will be correlated with coffee quality.Objective 2: Storm damage assessmentPlant phenotypes: plant height, biomass volume, canopy diameter, chlorophyll content, ratio of canopy area to total area, vegetation indices such as NDVI (normalized difference vegetative index), produce quality and others were identified as indicators to assess storm damage. These parameters will be field recorded from plantain fields prior to acquiring sensor data including storm damaged fields. Multi-spectral images and LIDAR scans of plantations will be acquired at the highest image resolutions. Analyzed LIDAR data and vegetation indices determined from multi-spectral images will be correlated with ground based data collected from cultivars and level of storm damage. An effort will be made to select a new plantation and follow its development throughout the length of the project for a complete description of plant development. Results from this work will help in development and adoption of precision agricultural practices to enhance yield and reduce cultivation costs. Most importantly, the historical data generated will be used to assess storm damage resulting from tropical storms and hurricanes.Objective 3: Forage yield mappingFollowing parameters: crop height, above ground biomass, plant density, and chlorophyll content will be recorded as indicators of forage yield and quality. A LIDAR sensor would be mounted on a drone to scan forage crop (quarterly). The sensor will scan the land area, at the highest resolution, at different crop growth stages. The scanning will begin before seeding for annual crops or crop sprouting for perennial crops and end before harvesting. The yield indicators, determined from sensed data, will be correlated with experimental data on forage quality and yield.

Progress 03/14/16 to 02/25/19

Outputs
Target Audience:Coffee, forage and citrus growers, and scientific community. Changes/Problems:Lack of interest on the research topic among the agricultural students. Lack of exposure to computer programming courses and level of expertize needed, appears to be the main cause for the lack of interest. It is recommended to attract graduate students with agricultural engineering background and programming skills. What opportunities for training and professional development has the project provided?Five undergraduate students and one graduate students were trained in acquiring and analyzing images of agricultural crops form a unmanned aerial vehicle. How have the results been disseminated to communities of interest?Three oral presentations and one poster presentation were made. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? 1. Major activities completed: a) Coffee characterization: Images of a coffee farm in Jayuya, PR and coffee crop characteristics were collected and maps generated. b. Storm damage assessment: Images of a citrus orchard were acquired and field data on citrus plant characteristics were also acquired. c. Forage yield sensing: Assessment of suitability of precision agricultural technologies for adoption in Puerto Rican dairy and forage industry. 2. Specific objectives met: All three specific objectives were met with preliminary information to strengthen future research proposals in adopting precision agriculture technologies in Puerto Rico. 3. Significant results: Preliminary data analysis showed that the measured citrus plant parameters and calculate parameters showed good correlation. The average errors in measurements were 22.4%, 14.5% and 47.5% for the plant diameter, plant height and plant canopy volume. Plant health indices were investigated and differences were found in disease severity between both the supplemental nutritional programs and the granular fertilization soil application. A manuscript for a scientific journal was prepared and submitted.

Publications

Type: Conference Papers and Presentations Status: Published Year Published: 2018 Citation: Mathanker, S. K. 2018. Hay-forage yield sensing for a sustainable dairy industry in Puerto Rico. 42nd Annual Meeting of the Puerto Rico Society of Agricultural Sciences (Sociedad Puertorrique�a de Ciencias Agr�colas, SOPCA). November 16, 2018. Aguadilla, Puerto Rico.

Progress 10/01/17 to 09/30/18

Outputs
Target Audience:Scientific community. Changes/Problems:Lack of interest on the research topic among the agricultural students. Lack of exposure to computer programming courses and level of expertize needed, appears to be the main cause lack of it. It is recommended to attract graduate students with agricultural engineering background. What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest?One oral presentation was made, and one journal article manuscript is submitted for publication. What do you plan to do during the next reporting period to accomplish the goals?One oral presentation in a scientific meeting.

Impacts
What was accomplished under these goals? . Major activities completed: A manuscript for a scientific journal was prepared and submitted. 2. Specific objectives met: Part of the objective 2 (Storm damage assessment) were met. 3. Significant results: Submission of a journal article manuscript.

Publications

Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Mathanker, S. K., A. M. Pagan-Lopez, C. Estevez-Jensen, & A. Beale. 2017. Determination of citrus plant parameters from UAV images. 15th National Meeting of Remote Sensing and GIS in Puerto Rico, December 08, 2017. Mayaguez, Puerto Rico USA.
Type: Journal Articles Status: Under Review Year Published: 2017 Citation: Mathanker, S. K., C. Estevez-Jensen, A. M. Pagan-Lopez, L. R. Perez-Alegria, and A. Beale. Comparison of field observed and calculated citrus plant parameters from UAV images. The Journal of Agriculture of the University of Puerto Rico.

Progress 10/01/16 to 09/30/17

Outputs
Target Audience:Scientific community and citrus growers. Changes/Problems:Citrus was selected as a model crop for storm damage assessment instead of plantain because the experiments of the relevant Co-PI have been concluded. What opportunities for training and professional development has the project provided?Two undergraduate students and one graduate student were trained on acquiring and analyzing images of agricultural crops. How have the results been disseminated to communities of interest?One oral presentation and one poster presentation were made. What do you plan to do during the next reporting period to accomplish the goals?Procured multi-spectral camera will be used to acquire multi-spectral images of a citrus experiment for correlating field and lab parameters with the calculated parameters from the multi-spectral images.

Impacts
What was accomplished under these goals? 1. Major activities completed: Plant parameters from citrus images were correlated with the field observed plant parameters. A multi-spectral camera has been procured and integrated with a drone. 2. Specific objectives met: Part of the objective 2 (Storm damage assessment) were met. 3. Significant results: The observed plant parameters correlated well with the determined plant parameters from the color UAV images. The average errors in measurements were 22.4%, 14.5% and 47.5% for the plant diameter, plant height and plant canopy volume. Plant health indices were investigated and differences were found in disease severity between both the supplemental nutritional programs and the granular fertilization soil application.

Publications

Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Mathanker, S. K., C. Estevez-Jensen, & A. Beale. 2017. Sensing and UAV technologies to assess citrus greening in Puerto Rico. 5th International Research Conference on Huanglongbing (HLB), March 14-17, 2017. Orlando, Florida USA.

Progress 03/14/16 to 09/30/16

Outputs
Target Audience:Scientific community and a coffee grower. Changes/Problems:Citrus was selected as a model crop for storm damage assessment instead of plantain because the experiments of the relevant Co-PI have been concluded. What opportunities for training and professional development has the project provided?Three undergraduate students were trained on acquiring and analyzing images of agricultural crops taken from a unmanned aerial vehicle. How have the results been disseminated to communities of interest?An oral presentation based on project results was made. What do you plan to do during the next reporting period to accomplish the goals?Remaining equipment will be acquired. A forage crop's images will be acquired and analyzed. Furthermore, coffee and citrus image analysis will be further developed.

Impacts
What was accomplished under these goals? 1. Major activities completed: a) Coffee characterization: Images of a coffee farm in Jayuya, PR and coffee crop characteristics were collected. Figure 1 shows selected images of the farm. Image stitching and incorporation of related information is under progress. b. Storm damage assessment: Images of a citrus orchard were acquired and field data on citrus plant characteristics were also acquired. [Citrus was selected as a model crop for storm damage assessment instead of plantain because the experiments of the relevant Co-PI have been concluded.] 2. Specific objectives met: Part of the objective 1 (Coffee characterization) and objective 2 (Storm damage assessment) were met. 3. Significant results: Preliminary data analysis showed that the measured citrus plant parameters and calculated parameters showed good correlation.

Publications

Type: Conference Papers and Presentations Status: Published Year Published: 2016 Citation: Mathanker, S. K., and L. R. Perez-Alegria. 2016. Image processing applications for plantains in Puerto Rico. 62nd Meeting. Inter-American Society for Tropical Horticulture. Rinc�n Beach Resort, A�asco, PR.