Giorgos Mountrakis, Ph.D.
Founder & Director | Intelligent Geocomputing Laboratory
Professor | Dept. of Environmental Resources Engineering | State University of NY College of Environmental Science and Forestry
Associate Editor | ISPRS Journal of Photogrammetry and Remote Sensing

Multiple Ph.D. Assistantships available for August 2024

Deep Learning in Remote sensing

The Landsat archive, with a multi-decadal global coverage is a prime candidate for deep learning classification methods due to the large data volume. We used 21 million labeled annual temporal sequences to compare deep learners with traditional methods. Results indicate substantial classification improvements of deep learning methods (DLMs) over the RF. These improvements are more pronounced on challenging to classify pixels in heterogenous areas. RF classification accuracy reaches about 70% on average, while DLMs are at 86%-95% range, depending on model architecture. See more here.

Remote sensing research highlights

Creation of pixel-based accuracy maps for RS classification

Detailed accuracy maps communicate effectively limitations of RS classification products. Using interpolation methods in the spectral space we demonstrate that ...
Read More

Fifteen years of published land cover classifiers

A meta-analysis covered fifteen years of published works in the top five RS journals. We identified the added value of ...
Read More

Review of global and regional land cover products

Proliferation of satellite data has resulted in a wide range of large scale land cover products. In this review we ...
Read More

Support Vector Machines in RS problems

Support Vector Machines are a promising family of classifiers due to their low requirements of training samples, fast training times ...
Read More

RS multi-step classification using intermediate results

A new RS classification algorithm is proposed where classification takes place in multiple steps. Initially, easy to classify pixels are ...
Read More

Accurate ground detection in waveform LiDAR signals

Waveform LiDAR data are frequently used to analyze vegetation vertical structure. The majority of LiDAR products (e.g. height, biomass) are ...
Read More

Biomass estimation using waveform LiDAR signals

Waveform LiDAR offer considerable benefits in large scale biomass estimation. A new algorithm is proposed that uses information from the ...
Read More

A novel multi-scale regression and classification method

Typically neural network algorithms use either large (e.g. sigmnoidal) or local (e.g. Gaussian) activation functions. Here we propose a new ...
Read More

Interdisciplinary research highlights

Forest attrition is higher in the Western US, in public and rural lands

Analysis of satellite data in the 1990s identified that the average distance to forest in ...
Read More

High urban land consumption linked to lower African American population, higher poverty rate and lower income per capita

Study used satellite data to look at developed land consumption across the continental US. Distinct ...
Read More

Hurricane Katrina flooding disproportionately affected African American neighborhoods

Integration of satellite images with Census data studied the flooding during hurricane Katrina in 2005 ...
Read More

Urban growth prediction models: Challenges moving forward

UGPMs have allowed us to understand the mechanisms behind urban sprawl. This review goes beyond ...
Read More

Wildlife vehicle collisions: A review of geographic models

Collisions of vehicles with wildlife pose significant dangers for humans and animals. This review examines ...
Read More

News highlights

2023. Apr: Special Issue on Deep Leaning on the Landsat Archive Announcement

2022. Jan: Grant from SRC Inc to develop deep learning classifiers for remote sensing

2021. Mar: Grant from NASA to develop a rewilding decision support environment for tortoises in the Galapagos.

2020. Mar: Grant from the Healthy Water Solutions Center of Excellence to incorporate satellite measurements in water quality assessments.
2019. Dec: Grant from Onondaga County to monitor harmful algal blooms using satellite observations.
2018. Sept: Promoted to Professor.
2017. Feb: NY Times and Washington Post covered our forest attrition work.
2015. May: SUNY ESF Exemplary Researcher Award Announcement
2015. Jan: New grant from NASA to study Grazing Systems in the Altai Mountain Transboundary Zone.
2015. Jan: Associate Editor for the ISPRS journal.
2014. Mar: New grant from USDA McIntire-Stennis Program to study uncertainty in satellite-derived forest maps.
2013. May: Promoted to Associate Professor with tenure.
2012. Sep: Received the Excellent Reviewer Award by ISPRS.
2012. May: Invited Keynote Speaker at the 32nd EARSeL Symposium (European RS).
2011. Sep: An educational paper on inquiry-based learning in RS courses was accepted.
2011. Apr: Students presented the space balloon project at the National Conference on Undergraduate Research.
2010. Nov: A review paper on SVM methods was accepted in the ISPRS journal.
2010. Jun: After invitation attended UN FAO meeting in Rome, Italy.
2009. Aug: Full patent awarded by the USPTO on multi-scale neural networks.
2009. Jan: New grant from NASA to study spatiotemporal relationships between bird migration and climate change.
2008. Oct: Guest Editor for a PE&RS Special Issue on Artificial Intelligence in Remote Sensing.
2008. Aug: Recipient of NASA’s prestigious New Investigator Award to study urban dynamics.