Machine Learning & Sensing Lab

Machine Learning & Sensing Lab

The Machine Learning and Sensing Laboratory develops machine learning methods for autonomously analyzing and understanding sensor data.

We investigate and develop artificial intelligence, machine learning, pattern recognition, computational intelligence, signal processing, and information fusion methods for application to sensing. Applications we have studied include landmine and explosive object detection, automated plant phenotyping, sub-pixel target detection, and underwater scene understanding. We have developed algorithms for ground-penetrating radar, hyperspectral imagery, electromagnetic induction data, synthetic aperture SONAR, and minirhizotron imagery. 

Current Projects

FreshID: Fruits and Vegetables Quality Evaluation Using Hyperspectral Imaging System
In this effort, we are developing AI-based approaches to detect and predict vegetable “freshness” levels.
Funding Agency:
USDA NIFA and UF Research
Role: PI: T. Liu, UF Horticulture; Co-PI: A. Zare
Student: Xiaolei Guo
Dates: June 2021 – Current
Related Links: UF Article 01 2021 | UF Article 02 2021 | WCJB News Story and Video 2021 | Gainesville Sun Article 2021

Understanding pre-harvest and post-harvest factors that can help to mitigate aflatoxin in peanut
This project is investigating and working to identify both pre- and post-harvest risk factors for Aflatoxin develop in peanut production. The effort involves imaging to directly detect evidence of aflatoxins as well as advancing models to combine environmental and management factors to predict aflatoxin risk.
Funding Agency: Industry Sponsors
Role: PI: B. Tillman at Univ. Florida; Co-PI: A. Zare
Student: Khaled Hamad
Dates: February 2021 – Current

AI-HARVEST: Artificial Intelligence Hub for Agricultural Reporting and Verification of Ecosystem Services through Sensing Technologies
In this effort, we are developing hybrid- deep learning and crop modeling tools to measure and quantify ecosystem services provided by an agricultural system.
Funding Agency:
FDACS
Role: PI: K-Morgan; Co-PI: A. Zare
Students: Weihuang Xu, Satya Krishna, Ayesha Naikodi
Dates: February 2021 – Current

SiTS: Hyperspectral Signals in the Soil
In this effort, we will build hyperspectral cameras (which also collect information outside of the visible range collected by standard color cameras) that will be inexpensive, have an automated mechanism to move up and down minirhizotron tubes, and are paired with automated algorithms to process and understand the collected data.
Funding Agency:
 USDA NIFA
Role: PI: A. Zare
Student: Ritesh Chowdhry
Dates: January 2021 – Current
Related Links: UF Article 2020 | Horti Daily Article 2021 | WCJB News Story and Video #1 2021 | WCJB News Story and Video #2 2021 | Villages Daily Sun 2021 | Fox 13 Tampa Bay 2021 | Florida Insider 2021 | Independent 2021 | Reuters 2021

Hurricane effects on the distribution and management of plant invasions in coastal habitats
In this effort, we are using on-the-ground plant measurements and aerial hyperspectral sensing to evaluate the effects of hurricanes on Brazilian peppertree.
Funding Agency:
 Everglades National Park
Role: PI: L. Flory, UF Agronomy; Co-PI: A. Zare
Dates: Sept. 2020 – Current
Related Links: UF Article 2021

AEROFUSION-MLUQ: Mars Lander Aerodynamic Model Data Fusion using Machine Learning
In this effort, the team is developing machine learning tools to quantify uncertainty in planetary landing systems. The effort will advance sensitivity analysis, outlier identification, multi-sensor fusion, and associated uncertainty quantification on sensor analysis performed.
Funding Agency:
NASA
Role: PI: M. Lee at NASA; Co-PI: A. Zare
Student: Yury Lebedev
Dates: August 2020 – Current

Robust and Reversible Deep Nets for Synthetic Aperture Sonar Analysis
This project is developed automated outlier detection and competency awareness techniques for deep learning architectures with specific application to synthetic aperture sonar image analysis.
Funding Agency:
Naval Surface Warfare Center
Role: PI: A. Zare
Students: Matthew Cook, Connor McCurley
Dates: August 2020 – Current

Testing predictions of plant-microbe-environment interactions to optimize climate adaptation and improve sustainability in switchgrass feedstocks
This work will leverage a field-established genetic resource network (S-GENE) to deepen understanding of local adaptation and to identify beneficial traits, genes, and microbial associates that contribute to switchgrass productivity.
Funding Agency: DOE
Role: PI: T. Juenger at UT Austin; Co-PI: A. Zare
Dates: September 2020 – Current
Student: Yiming Cui
Related Links: UF Article 2020 | UT Article 2020

NSF Engineering Research Center for Internet of Things for Precision Agriculture
This effort is an NSF ERC focused on the development of Internet of Things (IoT) for Precision Agriculture and the infrastructure (including algorithm development) needed to effectively deploy IoT devices for precision agriculture. 
Funding Agency: NSF
Role: PI: C. Kagan at UPenn; Co-PI: A. Zare
Dates: Sept. 2020 – Current
Student: Khaled Hamad
Related Links: UF Article #1 2020 | UF Article #2 2020 | UF Article #3 2020 | UF Article #4 2021 | IoT4Ag Homepage| NSF Award Article 2020 | Purdue Article 2020 | UC Merced Article 2020 | WCJB News Story and Video 2020

Satellite estimation of Bahiagrass yield and crude protein
In this effort, satellite remote sensing imagery is being used to estimate parameters of interest in pasture used for cattle ranching.
Funding Agency:
Industry Partners
Role: PI: C. Wilson, UF Agronomy; Co-PI: A. Zare
Dates: May 2020 – Current
Students: 
Related Links: UF Article 2021 | Gainesville Sun News Story 2021

Superpixel Segmentation and Texture Feature Learning for Multi-Aspect Underwater Scene Understanding
In this project, we are developing segmentation and texture feature analysis methods for automated underwater scene understanding using synthetic aperture sonar.
Funding Agency: DOD
PI: A. Zare
Student: Connor McCurley
Dates: May 2020 – Current

MRA: Disentangling Cross-scale Influences on Tree Species, Traits, and Diversity from Individual Trees to Continental Scales
Trees are essential to ecosystems. They store carbon, reduce erosion, and serve as habitat for other species. The factors influencing trees, and the spatial scales at which they are managed, range from an individual tree to entire continents. Since there are approximately three trillion trees in the world collecting data on every tree over large areas is impossible using traditional methods. Therefore, it is necessary to use new technology to measure and describe individual trees over large geographic areas. This research will address this fundamental challenge by combining high resolution remote sensing data with field data on trees.
Funding Agency: NSF
Role: PI: E. White at Univ. Florida; Co-PI: A. Zare
Dates: Aug. 2019 – Current
Students: Ritesh Chowdhry, Meilun Zhou, Matt Wein
Related Links: Project Website

SENTINEL: SENsing Threats In Natural Environments using Ligand-receptor modules
Funding Agency: DARPA
Role: Co-PI; PI: D. Nusinow, Danforth Plant Science Center
Dates: September 2018 – Current

Coordinated Adaptive Phenotyping (CAPs) for Improving Soil Water Acquisition
Funding Agency: USDA NIFA
Role: Co-PI;  PI: B. Tillman at Univ. Florida
Dates: June 2018 – Current Related Links: UF|IFAS Article

Improved system assessment of aflatoxin risk utilizing novel data and sensing
Funding Agency: USDA NIFA
Role: Co-PI;  PI: D. Rowland at Univ. Florida
Dates: May 2018 – Current
Related Links: UF|IFAS Article

Adaptive Manifold Learning for Multi-Sensor Translation & Fusion given Missing Data
Funding Agency: Center for Big Learning
Role: PI: A. Zare and P. Gader
Dates: May 2018 – Current

Completed Projects

Environmentally-aware Feature Extraction/Selection and Classification of Underwater Objects in Synthetic Aperture Sonar Imagery for Mine Countermeasures
Funding Agency: Office of Naval Research
Role: Co-PI;  PI: J. Keller at Univ. of Missouri
Dates: May 2016 – Dec. 2021

What mind matters? Machine learning approaches to linking structural variation in the brain to individual differences in spatial behavior
Funding Agency: UF Research, AI Catalyst Funds
Role: Co-PI; PI: S. Weisberg, UF Psychology
Dates: Nov. 2020 – Dec. 2021
Related Links: UF Article 2021

Parasitic Nematode Identification with Deep Learning
Funding Agency: UF Research, AI Catalyst Funds
PI: A. Zare
Dates: Nov. 2020 – Nov. 2021
Related Links: UF Article 2020 | UF IFAS Article 2021 | UF Explore Article 2021 | Growing Produce Article 2021

Rays for Roots: Integrating Backscatter X-Ray Phenotyping, Modeling, and Genetics to Increase Carbon Sequestration and Resource Use Efficiency
Funding Agency:ARPA-E
PI: A. Zare
Dates: May 2017 – Aug 2021
Related Links: ARPA-E ROOTS Project Descriptions | ARPA-E Announced $70M in Funding | UF HWCOE Article | UF Research Awards

Cyber-Physical Systems Security through Robust Adaptive Possibilitistic Algorithms: a Cross Layered Framework
Funding Agency: NSF
Role: Co-PI; PI: A. Bretas at Univ. Florida
Dates: August 2018 – July 2021

Multi-Sensor Data Fusion for Buried Target Detection
Funding Agency: Army Research Office
PI: A. Zare
Dates: April 2017 – May 2021

CAREER: Supervised Learning for Incomplete and Uncertain Data
Funding Agency: National Science Foundation
PI: A. Zare
Dates: May 2014 – Apr. 2021
Related Links: Project Website | NSF Award Abstract | MU Article| MIZZOU Magazine Article

Climate Adaptation and Sustainability in Switchgrass: Exploring Plant-Microbe-Soil Interactions across Continental Scale Environmental Gradients
Funding Agency: Department of Energy
Role: Co-PI;  PI: T. Juenger at Univ. of Texas at Austin
Dates: August 2015 – March 2021
Related Links: UT Austin Article | UF HWCOE Article

Development of Competitive Analytic Research Methodologies with Universities, Volunteer Geospatial Information Methodologies GEOINT Curriculum Development and Scientific Exchange
Funding Agency: NGA
Role: Co-PI; PI: P. Gader at Univ. Florida
Dates: March 2018 – March 2021

Multi-Aspect Underwater Scene Understanding
Funding Agency: Office of Naval Research
PI: A. Zare
Dates: Apr. 2014 – Dec. 2020
Related Links: MU Article 2015 | MU Article 2016

Taking the next steps towards making drone data applicable: novel approaches for directly relating UAV images to peanut maturity, disease, and drought
Funding Agency: National Peanut Board
PI: A. Zare
Dates: Jan. 2019 – June 2020

Peanut Volatile Organic Compounds Analysis
Funding Agency: National Peanut Board
PI: A. Zare
Dates: Jan. 2019 – June 2020 
Related Links: Georgia Tech Article

Biodiversity Data from Insect Songs: New hardware and software for monitoring insect bioacoustics, and new opportunities for public outreach
Funding Agency: UFII & UFBI
Role: Co-PI; PI: B. Stucky at Univ. Florida
Dates:  Nov. 2018 – May 2020

Peanut seed quality research
Funding Agency: APSA
Role: Co-PI; PI: D. Rowland at Univ. Florida
Dates: Aug. 2018 – July 2019

Robust Hyperspectral Image Analysis via Computational Topology
Funding Agency: UFII
Role: Co-PI; PI: P. Bubenik at Univ. Florida
Dates: Aug. 2017 – Aug. 2018

Algorithm and Decision Support for Buried Object Detection
Funding Agency: Army Research Office
PI: A. Zare
Dates: October 2016 – April 2017

New Investigator Program Award: Functions of Multiple Instances for Hyperspectral Analysis
Funding Agency: National Geospatial-Intelligence Agency
Dates: September 30, 2014 – October 2017
PI: A. Zare
Note: Research conducted at Dr. Zare’s prior appointment at the Univ. of Missouri

Machine Learning Techniques for Handheld Subsurface Object Detection
Funding Agency: Army Research Office
PI: A. Zare
Dates: November 1, 2013 – October 2016
Note: Research conducted at Dr. Zare’s prior appointment at the Univ. of Missouri

Understanding root growth using X-ray CT Imaging to increase crop yields
Funding Agency: Mizzou Advantage
Role: Co-PI;  PI: S. Kovaleski
Dates: June 2014 – September 2016
Note: Research conducted at Dr. Zare’s prior appointment at the Univ. of Missouri

Adaptive Underwater Target Detection
Funding Agency: MU Research Board
PI: A. Zare
Dates: January 1, 2014 – May 2015
Note: Research conducted at Dr. Zare’s prior appointment at the Univ. of Missouri

Environmentally-Adaptive Target Recognition Systems
Funding Agency: Naval Surface Warfare Center
PI: A. Zare
Dates: April 2014 – May 2015
Note: Research conducted at Dr. Zare’s prior appointment at the Univ. of Missouri

Mathematical Models for Describing and Reasoning with Geographic and Human Cultural Features
Funding Agency: DOD
Role: Co-PI;  PI: J. Keller
Dates: October 2010 – December 2014
Note: Research conducted at Dr. Zare’s prior appointment at the Univ. of Missouri

Explosive Object Detection with Electromagnetic Induction Sensors
Funding Agency: DOD
PI: A. Zare
Dates: September 2011 – May 2014
Note: Research conducted at Dr. Zare’s prior appointment at the Univ. of Missouri

Probabilistic Hyperspectral and LIDAR Fusion
Funding Agency: DOD
Dates: October 1, 2010 – October 31, 2013
Note: Research conducted at Dr. Zare’s prior appointment at the Univ. of Missouri

Airborne Multispectral Imagery (MSI) and ground-based Ground Penetrating Radar (GPR) Fusion
Funding Agency: DOD
PI: A. Zare
Dates: September 2010 – October 2011
Note: Research conducted at Dr. Zare’s prior appointment at the Univ. of Missouri