The Votes Are In: See the Winner of the HxGN LIVE EDU Contest

Many of us look forward to the spring for warmer weather or nature’s renewal. In Hexagon’s Geospatial division, the spring brings with it the many submissions to our EDU Contest for HxGN LIVE, showing us how students and professors are using Hexagon software or solutions in their academic work. This year’s contest, which we announced in February, showed us how our academic colleagues are studying the dynamically changing Earth.

This exciting opportunity offers students and professors utilizing Hexagon technology a chance to win free registration to HxGN LIVE 2019 at the Venetian in Las Vegas, Nevada, USA, June 11-14. The prize also includes a session to present their research and discounted travel expenses.

With so many great submissions to choose from, we had difficulty choosing just one winner. But we did!

The Winning Project for the HxGN LIVE EDU Contest is…

Chris Hanni - HxGN LIVE EDU Contest Winner

We would like to congratulate Chris Hanni, a graduate student from South Florida University, for his winning submission to the contest! In his submission, Chris detailed how he is using Hexagon technology in his project, entitled, Assessing Palm Decline in Florida by using Advanced Remote Sensing with Machine Learning Technologies and Algorithms.

Here is Chris’ abstract:

Native palms, such as the Sabal palmetto, play an important role in maintaining the ecological balance in Florida. As a potential side-effect of modern globalization, new phytopathogens like Texas Phoenix Palm Decline (TPPD) have been introduced into forest systems that threaten native palms. This presents new challenges for forestry managers and geographers. Advances in remote sensing have assisted the practice of forestry by providing spatial metrics regarding the type, quantity, location, and the state of health for trees for many years.

Spatial details regarding the general palm decline in Florida were elucidated by using the tools found in Hexagon’s ERDAS IMAGINE software in conjunction with R classification programing packages. IMAGINE Spatial Modeler, a graphical interface for building and running geoprocessing workflows, streamlined the development processes for data inputs during the stepwise refinement masking, feature layer production, and zonal statistic generation. Rapid adjustments to defined processing areas expedited the model tuning process maximizing the ability to take advantage of new developments in deep learning constructs coupled with high resolution WorldView-2 multispectral/temporal satellite imagery and lidar point cloud data. A novel approach using TensorFlow deep learning classification, multiband spatial statistics and indices, data reduction, and stepwise refinement masking yielded a significant improvement over Random Forest classification in a comparison analysis.

The results from the TensorFlow deep learning were used to develop an Empirical Bayesian Kriging continuous raster as an informative map regarding palm decline zones using Normalized Difference Vegetation Index Change. The significance from this research, using the data inputs generated from IMAGINE Spatial modeler, showed a large portion of the study area exhibiting palm decline and provided a new methodology for deploying TensorFlow learning for multispectral satellite imagery. The data sets generated from the model can aid law makers, government agencies, forestry managers, and environmental advocacy groups regarding the severity of the decline.

Additionally, recommendations on future ERDAS IMAGINE machine/deep learning builds will be provided. Classification parameter accessibility, GPU acceleration, and performance feedback options will be discussed.

Congratulations Again, Chris!

In closing, we are always thrilled to learn how Hexagon technology is helping to solve real-world problems such as, in this case, discovering how much damage to palm trees is being caused by parasitic organisms. It is especially gratifying to see how cutting-edge technology such as TensorFlow deep learning classification can be incorporated into machine learning operators in IMAGINE Spatial Modeler.

Stay tuned to the Sensing Change Blog for more details on Chris’ winning project.

Chris will be presenting his work at HxGN LIVE 2019. Register for the conference to learn more about his project firsthand!

  • Become Remote Sensing Certified with BRS-Labs

    Brilliant Remote Sensing Labs is helping to make remote sensing education more accessible than ever. See how students are applying these skills across industries.

  • Recent Posts