Title: Post-Doctoral Researcher - Machine Vision/Remote Sensing Specialist
Location: Beltsville, MD USA
Post Doctoral Researcher, USDA-ARS BARC Sustainable Agricultural Systems Laboratory and Hydrology and Remote Sensing Laboratory, Beltsville, MD, University of Maryland, Environmental Science and Technology, College Park, MD, and North Carolina State University, Dept. of Crop and Soils.
Starting Date: As soon as possible
Salary: GS 11/12 DC Metro area; based on experience
This position provides remote sensing, geospatial statistics, and machine vision and learning support for a USDA-ARS team (Partnerships for Data Innovations project and Agricultural Collaborative Outcome System (AgCROS)), the Precision Sustainable Agricultural team (national network of sustainable agriculture scientists), and in partnership with Microsoft, Esri, and several other technology companies. The candidate will work with a team of applied agricultural scientists, technologists, and data scientists to assess cover crop and weed dynamics in agronomic cropping systems using a suite of remote sensing systems (satellite, airborne, drone, tractor-mounted, field deployed). Using machine vision and learning approaches, the candidate would assess cover crop and weed performance spatially to inform decision support tools, modeling efforts, and precision technologies. Additional work will include integration into the Microsoft FarmBeats platform for existing data streams produced by a collaborative team: direct measures of cash and cover crops (nutrients, biomass, yield), soils (sensor-based water content, EC, temperature measurements), remote sensing (satellites, drones), and proximal sensing (tractor-based LIDAR, radar, multispectral/hyperspectral cameras).
Duties and Responsibilities:
Support deployment, quality control, and analysis of a diverse set of remote sensing approaches to quantifying cover crop and weed performance in field crop production systems across a distributed network of researchers in the US. Lead development of insights around optimal linkage between sensing technology with plant performance criteria, conduct AI/ML analysis, and support engineering of work and data flow. Conduct geospatial analysis of cover crops that link a range of remote sensing resolutions with field destructive samples. Explore linkages between sensing systems for plant growth to process-based modeling approaches. Participate in decision support tool development.
The successful candidate will be exceptionally well-organized, a strong writer, a strong project manager, and have the ability to conduct and lead engineering research including image processing and computer vision for agriculture, hardware automation for outdoor environments, embedded systems, programming skills, and high throughput phenotyping. Candidates should have training and experience with computer vision algorithms and libraries, machine learning for pattern recognition, and hardware interfacing. Demonstrate experience building and applying imaging and analysis techniques to a wide range of problems in machine vision. Demonstrated ability to deliver scientific results and communicate machine learning concepts utilized in a cross functional environment. The candidate will deploy IoT technologies automating data acquisition and integration from a wide variety of sensors. The candidate will Identify, plan and execute on opportunities to use hardware and software to automate R&D operations and increase field efficiencies.
He/she will have demonstrated research productivity through publications in relevant refereed journals, and an existing record of, or strong potential for, successful grant procurement. It is essential that the incumbent conduct team-oriented research, exhibit exceptional leadership abilities, and demonstrate effective written and verbal communication skills.
Required job qualifications:
PhD in applied agricultural sciences, spatial statistics, landscape ecology, data science, machine learning, computer vision, remote sensing or related field
Ability to migrate data streams into and out of the ESRI GIS suite of products.
Classical machine learning and modern deep learning approaches
Familiarity with soil-landscape relationships, how soils influence plant growth, and soil survey databases available from NRCS.
Preferred job qualifications:
Working knowledge of user interface development systems used in software application development such as git, SQL, RStudio
Working knowledge of Microsoft Windows, Microsoft Windows Server, and Linux.
Working knowledge of cloud platforms (Azure, AWS or GEE).
Competence in server administration and basic cybersecurity of distributed systems.
Valid drivers license.
Must be able to successfully pass the background check process.
Questions regarding this position may be directed to:
Dr. Steven Mirsky, USDA-ARS Sustainable Agricultural Systems Laboratory, Beltsville, MD, [email protected], 240-304-9479
Key partners/mentors include:
Dr. Dean Hively, USGS stationed at USDA-ARS Hydrology and Remote Sensing Laboratory, Beltsville, MD, [email protected], 301-504-9031
Dr. Brian Needelman, Department of Environmental Science and Technology, University of Maryland, College Park, MD 20742, [email protected], 301-405-8227
Dr. S. Chris Reberg-Horton, Department of Crop and Soils, North Carolina State University, Raleigh, [email protected], 919 515-7597
All Contents © 1999-2019. The GIS Jobs Clearinghouse