Skip to main content
ARS Home » Pacific West Area » Maricopa, Arizona » U.S. Arid Land Agricultural Research Center » Plant Physiology and Genetics Research » Research » Publications at this Location » Publication #379041

Research Project: Enhancing Abiotic Stress Tolerance of Cotton, Oilseeds, and Other Industrial and Biofuel Crops Using High Throughput Phenotyping and Other Genetic Approaches

Location: Plant Physiology and Genetics Research

Title: Software design for image mapping and analytics for high throughput phenotyping

item Kim, James

Submitted to: Computers and Electronics in Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/2/2021
Publication Date: 11/10/2021
Citation: Kim, J.Y. 2021. Software design for image mapping and analytics for high throughput phenotyping. Computers and Electronics in Agriculture. 191. Article 106550.

Interpretive Summary: High throughput phenotyping (HTP) is highly demanding on breeding pipeline to identify superior varieties and genetic gains. Image-based HTP has been widely used for HTP, with advantages of large coverage and canopy extraction of morphological and spectral signature through image processing technology. However, algorithm development of image processing is challenging to meet cost-effectiveness and global consistency of customized sensors and platforms. User-friendly analytic software for image mapping and analytics for phenotyping (IMAP) was developed and graphically interfaced for the user to easily process and analyze the HTP data. The IMAP software offers GIS interface for the user to interchangeably create and use a shapefile and a region of interest by drawing a grid with the mouse click and drag. It also supports geometric and radiometric calibrations to address the geometric drift and spectral inconsistency under ambient illumination changes. The IMAP software allows the user to automate multiple tasks in a sequence by combining modular functions using a batch process. The IMAP software was coded in open source language Python and easily updated and combined with other functions for the future applications on various HTP platforms.

Technical Abstract: Phenotyping facilitates plant breeding, and image-based high throughput phenotyping (HTP) has been widely adopted to expedite a breeding pipeline with advantages of large coverage and image processing for morphological and spectral signature of the plant canopy. To meet the demand of a cost-effective and globally consistent HTP data processing pipeline, customized software was developed to process field data and images to extract plot-level metrics of plant phenotypes. Graphic user interface allowed the user to easily perform image processing, data visualization and batch processing. The algorithms included GIS interface for field and plot boundaries, geometric calibration to align the data with the plots, spectral calibration to accommodate ambient illumination change, geo-fencing within a field or plot boundary, and batch processes. Ground registration and spectral analysis in plant-level resolution were delivered and validated with ground and aerial truthing.