Earlier this year we hosted a Geospatial Education Contest offering students and professors a chance to present their research using Hexagon Geospatial technology to business leaders at our annual conference, HxGN LIVE.
We would like to spotlight one of our honorable mentions from the contest, Dr. Ariel Blanco, an Associate Professor of Geodetic Engineering at the University of the Philippines, for his innovative research using Hexagon Geospatial technology. Ariel’s research focuses on improving the present system for regulating sugarcane, as the Philippines is one of the leading producers of sugarcane in the world. The Sugar Regulatory Administration (SRA) regulates the sugar industry. They require accurate information on areas planted and the condition of canes at specified periods during the cropping cycles in order to estimate or forecast on sugar production and supply.
Ariel and the University of the Philippines, along with the SRA conducted a research and development project entitled “Development of the Sugar Regulatory Administration Yield Estimation System for Sugarcane (SRA YESS) Project” through the UP Training Center for Applied Geodesy and Photogrammetry.
The goal of this project was to provide the SRA with reliable data and information for the entire Philippines. To do this the SRA YESS Project had to address the challenges associated with the SRA’s current monitoring and estimation system, which heavily relied on the reports of field personnel and data submitted by the milling districts which implement sugarcane production. Data availability and quality are critical issues, potentially leading to unreliable estimates.
Ariel goes into more detail about his research below:
The estimation of yield is a lengthy process requiring various datasets including satellite images, crop measurements, meteorological data, and many others. The process pertaining to images includes image stacking, radiometric correction, atmospheric correction (using ATCOR2), image transformation, image classification, and derivation of indices. The YESS system is meant to speed up the monitoring and estimation processes.
To achieve this goal, data processing was automated using workflows implemented as models in ERDAS IMAGINE using the Spatial Modeller. The Yield-forming-to-Mature-Cane (YMC) model, for example, implements a three-stage workflow comprised of (1) Principal Component Analysis (PCA) to produce a 3-band PCA image, (2) Isodata Classification to classify the PCA image into different classes using a algorithm with predefined input values (e.g., for convergence threshold, range of classes, and minimum distance), and (3) Zonal Statistics to combine all the classes corresponding to mature/ripening canes only, compute the corresponding total areas, and visualize the mature/ripening canes areas. The IMAGINE Spatial Modeller enabled quick and operational use of workflows and models developed by the YESS Project.
We would like to congratulate Ariel and the UP-SRA YESS Team on their hard work and for their innovative use of Hexagon Geospatial technology.