TERRA-REF - sensor data

Quantifying Leaf Chlorophyll Concentration of Sorghum from Hyperspectral Data Using Derivative Calculus and Machine Learning

dlebauer — Wed, 15 Sep 2021 22:53:56 +0000

June 29, 2020

Bhadra, Sourav, Vasit Sagan, Maitiniyazi Maimaitijiang, Matthew Maimaitiyiming, Maria Newcomb, Nadia Shakoor, and Todd C. Mockler. "Quantifying leaf chlorophyll concentration of sorghum from hyperspectral data using derivative calculus and machine learning." Remote Sensing 12, no. 13 (2020): 2082. https://www.mdpi.com/2072-4292/12/13/2082

Bhadra, Sourav, Vasit Sagan, Maitiniyazi Maimaitijiang, Matthew Maimaitiyiming, Maria Newcomb, Nadia Shakoor, and Todd C. Mockler

remotesensing-12-02082-v2.pdf

Multi-Resolution Outlier Pooling for Sorghum Classification

dlebauer — Wed, 15 Sep 2021 17:38:31 +0000

field scanner

reference data

sensor data

algorithms

June 19, 2021

This paper introduces the Sorghum-100 dataset, a large dataset of RGB imagery captured by the TERRA-REF field scanner as well as a new and more powerful approach to identifying which plant variety (cultivar) is in a particular image.

The Sorghum-100 dataset was used in the Sorghum-100 Kaggle competition, which explains:

The Sorghum-100 dataset is a curated subset of the RGB imagery captured during the TERRA-REF experiments, labeled by cultivar. This data could be used to develop and assess a variety of plant phenotyping models which seek to answer questions relating to the presence or absence of desirable traits (e.g., "does this plant exhibit signs of water stress?''). In this contest, we focus on the question: ``What cultivar is shown in this image?''

Chao Ren, Justin Dulay, Gregory Rolwes, Duke Pauli, Nadia Shakoor, Abby Stylianou

Ren_Multi-Resolution_Outlier_Pooling_for_Sorghum_Classification_CVPRW_2021_paper.pdf

Chlorophyll fluorescence imaging captures photochemical efficiency of grain sorghum (Sorghum bicolor) in a field setting

dlebauer — Wed, 15 Sep 2021 17:28:24 +0000

trait data

field scanner

reference data

sensor data

algorithms

Aug. 10, 2020

Herritt et al 2020 demonstrate the PSII camera's ability to capture PSII fluorescence by treating leaves with a chemical that inhibits photosynthesis (DMCU) and observing differences.

Herritt, M.T., Pauli, D., Mockler, T.C. et al. Chlorophyll fluorescence imaging captures photochemical efficiency of grain sorghum (Sorghum bicolor) in a field setting. Plant Methods 16, 109 (2020). https://doi.org/10.1186/s13007-020-00650-0

Herritt, Matthew T., Duke Pauli, Todd C. Mockler, and Alison L. Thompson

Herritt2020_Article_ChlorophyllFluorescenceImaging.pdf

What Does TERRA-REF's High Resolution, Multi Sensor Plant Sensing Public Domain Data Offer the Computer Vision Community?

dlebauer — Wed, 18 Aug 2021 19:26:27 +0000

reference data

sensor data

algorithms

Oct. 11, 2021

Our paper "What Does TERRA-REF's High Resolution, Multi Sensor Plant Sensing Public Domain Data Offer the Computer Vision Community?" was accepted for the 7th workshop on Computer Vision in Plant Phenotyping and Agriculture, which will be held Oct 11 at ICCV 2021. The goal of the paper is to introduce TERRA REF data to the machine learning and computer vision communities by providing a high level overview and a description of some computer vision and machine learning problems that can be addressed with this dataset.

The paper is available on Arxiv [2107.14072] prior to publication of the conference proceedings.

David LeBauer, Max Burnette, Noah Fahlgren, Rob Kooper, Kenton McHenry, Abby Stylianou

53.pdf

Data-Driven Artificial Intelligence for Calibration of Hyperspectral Big Data

dlebauer — Wed, 18 Aug 2021 17:57:52 +0000

field scanner

reference data

software

sensor data

algorithms

July 15, 2021

This paper describes the pipeline used to produce calibrated hyperspectral images from the two hyperspectral cameras deployed under field conditions. The pipeline implements radiometric calibration, reflectance calculation, bidirectional reflectance distribution function (BRDF) correction, soil and shadow masking, and image quality assessment. This was a true tour de force!

Calibrating these cameras was one of the project's most challenging efforts - the light environment is changing, and impacted not only by the sun but also by the big white box that both reflects and shades, the complex canopy, the slow moving camera and timing of calibration, not to mention broken cameras and the variety of radiometers and calibration targets that were put to use. The issues on GitHub include an interesting record of the discussions, challenges, and iterations of different approaches that we took over the years.

Sagan, Vasit, et al. "Data-Driven Artificial Intelligence for Calibration of Hyperspectral Big Data." IEEE Transactions on Geoscience and Remote Sensing (2021).

Vasit Sagan, Maitiniyazi Maimaitijiang, Sidike Paheding, Sourav Bhadra, Nichole Gosselin, Max Burnette, Jeffrey Demieville, Sean Hrtling, David LeBauer, Maria Newcomb, Duke Paul, Kyle Peterson, Nadia Shakoor, Abby Stylianou, Charles Zender, Todd Mockler

sagan2021dda_small.pdf

Learning to Correct for Bad Camera Settings in Large Scale Plant Monitoring

dlebauer — Tue, 18 Jun 2019 18:26:57 +0000

computing pipeline

sensor data

algorithms

June 18, 2019

Zongyang presented his research and implementation on an algorithm to correct image exposure at the CVPR 2019 workshop on "Computer Vision Problems in Plant Phenotyping".

Abstract: In large scale, automated image capture systems,incorrect camera settings can lead to images that are completelyuseless or which break assumptions made by image analysisalgorithms, but there may also be sufficient data to learn to automatically correct bad data. We consider the specific problem of over- and under-saturated images in large scale plant growthmonitoring, propose a generative approach that addresses thesesaturation issues for the calculation of plant canopy cover, and suggest future areas of research in this problem domain.

The source code for this algorithm is released under the BSD 3-Clause Open Source License and can be found in the TERRA REF Stereo RGB repository on GitHub.

Zongyang Li, Abby Stylianou, and Robert Pless

Li et al 2019 PDF

UAV-Based high resolution thermal imaging for vegetation monitoring and plant phenotyping using ICI 8640 P, FLIR Vue Pro R 640 and thermoMap Cameras

dlebauer — Tue, 05 Feb 2019 22:22:00 +0000

unmanned arial vehicles

trait data

sensor data

Feb. 7, 2019

The growing popularity of Unmanned Aerial Vehicles (UAVs) in recent years, along with decreased cost and more accessibility of both UAVs and thermal imaging sensors, has led to the widespread usage of this technology, especially for precision agriculture and plant phenotyping. There are several thermal camera systems in the market becoming more available at a low cost. However, their efficacy and accuracy in various applications has not been tested. In this study, three commercially available UAV thermal cameras, including ICI 8640 P-series camera (Infrared Cameras Inc., USA), a FLIR Vue Pro R 640 Thermal Camera (FLIR Systems, USA), and thermoMap (senseFly, Switzerland) have been tested and evaluated for their potential for forest monitoring, vegetation stress detection and plant phenotyping. Mounted on multi-rotor systems, these sensors were simultaneously flown over different experimental sites located in St. Louis, Missouri (forest environment), Columbia, Missouri (plant stress detection and phenotyping), and Maricopa, Arizona (high throughput phenotyping). Thermal camera datasets were calibrated using procedures that utilize a blackbody, ground thermal targets, emissivity and atmospheric correction. A suite of statistical analysis including analysis of variance test (ANOVA), correlation analysis between sensor temperature and plant biophysical and biochemical traits, and genotype heritability test was utilized to examine the sensitivity of the cameras against selected plant phenotypic traits and water stress detection. Our results showed that (1) UAV-based thermal imaging is a viable tool in precision agriculture and (2) all these systems are comparable in terms of their efficacy for plant phenotyping. Overall, accuracy when compared against field measured ground temperature and estimating power of plant biophysical and biochemical traits ICI 8640 P-series performed better than the other two cameras, which was followed by FLIR Vue Pro R 640 and thermoMap. Our results demonstrated that all three UAV thermal cameras provide useful temperature data for precision agriculture, ICI 8640 P-series being the best among the three systems compared; but cost wise, FLIR Vue Pro R 640 is the most affordable than other two cameras compared providing a cheaper option for a wide range of applications.

Sagan, V., Maimaitijiang, M., Sidike, P., Eblimit, K., Peterson, K.T., Hartling, S., Esposito, F., Khanal, K., Newcomb, M., Pauli, D., Ward, R., Fritschi, F., Shakoor, N., Mockler, T.

remotesensing-11-00330-v2.pdf

TERRA-REF Data Processing Infrastructure

dlebauer — Thu, 31 Jan 2019 16:35:16 +0000

trait data

standards

field scanner

computing pipeline

analysis workbench

reference data

software

sensor data

algorithms

July 22, 2018

The Transportation Energy Resources from Renewable Agriculture Phenotyping Reference Platform (TERRA-REF) provides a data and computation pipeline responsible for collecting, transferring, processing and distributing large volumes of crop sensing and genomic data from genetically informative germplasm sets. The primary source of these data is a field scanner system built over an experimental field at the University of Arizona Maricopa Agricultural Center. The scanner uses several different sensors to observe the field at a dense collection frequency with high resolution. These sensors include RGB stereo, thermal, pulse-amplitude modulated chlorophyll fluorescence, imaging spectrometer cameras, a 3D laser scanner, and environmental monitors. In addition, data from sensors mounted on tractors, UAVs, an indoor controlled-environment facility, and manually collected measurements are integrated into the pipeline. Upt to two TB of data per day are collected and transferred to NCSA at the University of Illinois where they are processed.

In this paper we describe the technical architecture for the TERRA-REF data and computing pipeline. This modular and scalable pipeline provides a suite of components to convert raw imagery to standard formats, geospatially subset data, and identify biophysical and physiological plant features related to crop productivity, resource use, and stress tolerance. Derived data products are uploaded to the Clowder content management system and the BETYdb traits and yields database for querying, supporting research at an experimental plot level. All software is open source under a BSD 3-clause or similar license and the data products are open access (currently for evaluation with a full release in fall 2019). In addition, we provide computing environments in which users can explore data and develop new tools. The goal of this system is to enable scientists to evaluate and use data, create new algorithms, and advance the science of digital agriculture and crop improvement.

Burnette, Willis, Kooper, Maloney, Ward, Shakoor, Newcomb, Rohde, Fahlgren, Sagan, Sidike, Terstriep, LeBauer

Burnette et al 2018 Article