PROGRAM: Schedule and Abstracts
MONDAY
- 8:30 Registration
- 9:00 Opening Session
- 9:15 Machine Learning and Remote Sensing for Wildfire and Spatial Risk Management
- 10:30 Coffee break
- 10:45 Machine Learning and Remote Sensing for Wildfire and Spatial Risk Management
- 12:30 Lunch
- 14:00 Landsat-derived global rainfed and irrigated-cropland product @ 30 m (LGRIP30)
- 15:30 Coffee break
- 15:45 New Generation of Hyperspectral Remote Sensing Data to Usher a New Era in Satellite Remote Sensing of Agriculture
- 17:30 Icebreaker
TUESDAY
- 9:00 A comprehensive tool for analyzing and processing hyperspectral data
- 10:30 Coffee break
- 10:45 A comprehensive tool for analyzing and processing hyperspectral data
- 12:30 Lunch
- 14:00 A comprehensive tool for analyzing and processing hyperspectral data
- 15:30 Coffee break
- 15:45 EO for Mapping Shallow Water Marine Habitats
- 17:30 Class ends / Poster Session
WEDNESDAY
- 9:00 Field Trip: Krka National Parkwith scheduled drone and LiDAR data collection activities.
- 12:30 Lunch
- 14: 30 Field Trip: Krka National Park with scheduled drone and LiDAR data collection activities.
- 17:30 Return to Campus
THURSDAY
- 9:00 Basics on SAR with a practical session on preprocessing of SAR image
- 10:30 Coffee break
- 10:45 Basics on SAR with a practical session on preprocessing of SAR image
- 12:30 Lunch
- 14:00 Basics on SAR with a practical session on preprocessing of SAR image
- 15:30 Coffee break
- 15:45 Basics on SAR with a practical session on preprocessing of SAR image
- 17:30 Class ends
- 19:30 Split @ night – Guided walking tour
- ~21:00 Social dinner (0ptional)
FRIDAY
- 9:00 UAS LiDAR for Landslide Slope Monitoring and Change Detection
- 10:30 Coffee break
- 10:45 UAS LiDAR for Landslide Slope Monitoring and Change Detection
- 12:30 Lunch
- 14:00 A hands-on course focused on handheld LiDAR data collection and processing
- 15:30 Coffee break
- 15:45 A hands-on course focused on handheld LiDAR data collection and processing
- 18:00 Class ends / Poster Session
SATURDAY
- 9:00 A hands-on course focused on handheld LiDAR data collection and processing – cont.
- 10:30 Coffee break
- 10:45 OBIA: Point Clouds and Object primitives in theory and practice
- 12:30 Lunch
- 14:00 OBIA: Point Clouds and Object primitives in theory and practice
- 15:30 Lectures end / Adjourn the event
Expand each day for more information !
TOPICS AND PRESENTERS!
• Machine Learning and Remote Sensing for Wildfire and Spatial Risk Management (Dr. Ljiljana Seric, FESB, Croatia)
• Landsat-derived global rainfed and irrigated-cropland product @ 30 m (LGRIP30) (Dr. Prasad Thenkabail, USGS, USA)
• New Generation of Hyperspectral Remote Sensing Data to Usher a New Era in Satellite Remote Sensing of Agriculture (Dr. Prasad Thenkabail, USA)
• ARTMO Toolbox: A comprehensive tool for analyzing and processing hyperspectral data (Dr. Jochem Verrelst, University of Valencia, Spain).
• Drone/UAV/UAS: UAS LiDAR for Landslide Slope Monitoring and Change Detection (Prof. Donna Delparte, Idaho State University, USA)
• LiDAR: A hands-on course focused on handheld LiDAR data collection and processing (Dr. Ivan Balenovic, Croatian Forest Research Institute, and Luka Zalovic, Geo-Centar, Croatia)
• Object Based Image Analysis (OBIA) with eCognition: EO for Mapping Shallow Water Marine Habitats / Point Clouds and Object primitives in theory and practice (Branimir Radun and Dr. Ivan Tomljenovic, Oikon Ltd., Croatia)
• SAR – Basics on SAR with a practical session on preprocessing of SAR image (Michele Claus, eurac research, Italy),
• Field Trip: Explore Krka National Park and the charming town of Skradin, with scheduled drone and LiDAR data collection activities.
2025 – BRING YOUR POSTER!
Discuss your project and poster with remote sensing experts.
ABSTRACTS:
Landsat-derived global rainfed and irrigated-cropland product @ 30 m (LGRIP30)
Climate variability and ballooning populations are putting unprecedented pressure on agricultural croplands and their water use, which are vital for ensuring global food and water security in the twenty-first century. In addition, the COVID-19 pandemic, military conflicts, and changing diets have added to looming global food insecurity. Therefore, in this lecture, we will focus on the release of the Landsat-derived global rainfed and irrigated-cropland product @ 30 m (LGRIP30). Importance of mapping irrigated and rainfed croplands cannot be overemphasized. Both irrigated and rainfed areas are water guzzlers since about 80-90% of all human water use goes towards producing food. Irrigated areas consume blue water. That is the water delivered to farms through irrigation systems either from surface water (e.g., lakes, reservoirs, tanks, river diversions) or ground water resources (e.g., tube wells, open wells). Rainfed areas consume green water. Currently, the highest resolution irrigated and rainfed cropland products derived from remote sensing are at nominal 1-km spatial resolution derived by fusing several existing products. The LGRIP30 cropland product was generated using Landsat-8 30m time-series data, multiple supervised and unsupervised Machine learning algorithms (MLAs) such as random forest, support vector machines, decision trees, ISOCLASS clustering, and spectral matching techniques. Landsat 8 time-series surface reflectance product that is already available on the Google Earth Engine (GEE) was the primary data used in the study.
Machine Learning and Remote Sensing for Wildfire and Spatial Risk Management
This interactive workshop offers hands-on experience in integrating artificial intelligence (AI) and remote sensing for wildfire prediction, analysis, and broader spatial risk management. Participants will gain a solid foundation in satellite platforms, learning to interpret key satellite products such as hotspots, burned area mapping, and vegetation moisture indices, essential for assessing environmental risks. Through hands-on exercises, attendees will explore machine learning techniques for pixel and object classification, fuel classification, and moisture regression—vital for understanding fire dynamics and risk zones. Practical exercises will also include fire spread behavior reconstruction, allowing participants to gain critical insights into wildfire propagation patterns. With real-world case studies and interactive sessions, this workshop equips attendees with the technical skills to harness satellite data and AI for comprehensive management in dynamic and uncertain environments. These skills are invaluable not only for wildfire managers but also for professionals across sectors involved in spatial risk management, including disaster response, environmental monitoring, and infrastructure resilience planning. By the end of the workshop, participants will have the practical skills and insights needed to utilize AI and satellite data for proactive risk management in dynamic and uncertain environments.
A hands-on course focused on handheld LiDAR data collection and processing
Rapid progress in sensor miniaturization has led to the development of lightweight and highly mobile personal laser scanning (PLS) systems that can be carried or held by a person. The emergence and availability of next-generation PLS systems, with significantly improved features (e.g., scanning range and rate, measurement accuracy) of LiDAR sensors, have recently sparked more intensive research into their application in forest inventory, particularly for estimating key tree attributes.
This interactive workshop will begin with a general overview (presentation) of the state-of-the-art hand-held PLS technology, as well as an introduction to the main principles of data collection and processing. Following the introduction, students will participate in field data collection using a high-end PLS instrument (Faro Orbis) in a vegetated area. The workshop will then continue with a hands-on session where students will perform the typical steps of a PLS LiDAR data processing workflow on their own computers using the previously collected data and provided software.
First, students will ‘preprocess’ the collected data (e.g., converting raw data to .las/.laz format, point cloud registration, and colorization) using Faro SCENE software. In the next step, students will learn how to use LiDAR360 software for further processing (point cloud classification, normalization, segmentation, etc.) and extraction of key tree attributes (diameter at breast height, tree height, crown dimensions).
OBIA using eCognition: EO for Mapping Shallow Water Marine Habitats
This workshop will focus on the use of Earth Observation (EO) data in mapping shallow water marine habitats. Participants will gain insight into the importance of shallow water marine habitats, challenges associated with using optical satellite data for such applications, including issues such as the effects of the water column, turbidity, and spatial resolution limitations. The session will provide practical solutions to these challenges by demonstrating the fusion of different image processing approaches.
Through real-world examples, participants will explore how these methods are applied to classify and delineate benthic habitats of the Adriatic Sea, including seagrass meadows and other key ecological features. Hands-on activities using tools like eCognition will guide attendees in applying classification rules to the marine environment, integrating multiple data sources, and performing accuracy assessments. These exercises will equip participants with practical skills to address real-world challenges and ensure robust and transferable habitat mapping results.
New Generation of Hyperspectral Remote Sensing Data to Usher a New Era in Satellite Remote Sensing of Agriculture
We are entering a new era of remote sensing with data acquisition from several spaceborne hyperspectral sensors or imaging spectroscopy. These include data from current sensors such as the: 1. German Deutsches Zentrum fur Luftund Raumfahrt (DLR’s) Earth Sensing Imaging Spectrometer (DESIS) sensor onboard the International Space Station (ISS), 2. Italian Space Agency (ASI) PRISMA (Hyperspectral Precursor of the Application Mission), and 3. German DLR’s Environmental Mapping and Analysis Program (EnMAP). They also include upcoming hyperspectral sensors such as the: 1. Planet Labs PBC’s (Public Benefit Corporation) two hyperspectral sensors called Tanager in 2023, and 2. NASA’s hyperspectral sensor Surface Biology and Geology (SBG) mission. Further, we already have over 83,000 hyperspectral images of the Earth acquired from NASA’s Earth Observing-1 (EO-1) Hyperion that are freely available to anyone through the U.S. Geological Survey’s (USGS) data archives. These suites of sensors acquire data in 200+ hyperspectral narrowbands (HNBs) in 2.55 to 12 nm bandwidth, either in 400-1000 or 400-2500 nm spectral range with SBG also acquiring data in the thermal range. The HNBs provide data as “spectral signatures” in stark contrast to “a few data points along the spectrum” provided by multispectral broadbands (MBBs) such as the Landsat satellite series. In this talk, we will demonstrate a few key advantages of HNB data relative to MBB data. First, we will discuss the five broad pillars of hyperspectral data analysis in studies pertaining to agriculture and vegetation. These are:
- Full spectral analysis (FSA) utilizing the entire spectrum.
- Optimal hyperspectral narrowband analysis (OHNBs) utilizing optimal bands.
- Hyperspectral vegetation indices (HVIs) developing two-band and multi-band HVIs.
- Classification of crop type classifications, and species type classifications. and
- Modeling physically based models, and statistical models.
An exhaustive discussion of these five pillars will be presented with concrete examples of how they are implemented and the strengths of HNBs as opposed to MBBs based on our comprehensive research over the years.
UAS LiDAR for Landslide Slope Monitoring and Change Detection
UAS LiDAR enables cost-effective temporal mapping of landslide slopes using high resolution datasets. In this workshop, students will learn the entire UAS workflow from mission planning and data processing to analysis and error estimation. Students will gain hands-on experience in analyzing UAS LiDAR data to detect change in slope stability. By using point cloud analysis, students will determine 3D slope changes and identify susceptible slopes. LAStools will be used during this hands-on session. LAStools are popular for their blazing speed and high productivity. The software combines robust algorithms with efficient I/O and memory-efficient management to achieve high throughput for datasets containing billions of points. The software runs on Windows and Linux. It has deep market-penetration and is widely used across industry, government agencies, research labs, and educational institutions. The ability to script the modules makes them ideal for use on web servers or in the cloud” (Lidar Magazine, 2023). rapidlasso GmbH was founded in 2012 by lidar pioneer Dr. Martin Isenburg. The company’s flagship product, the LAStools software suite, is a collection of 56 highly efficient, batch-scriptable, multicore command-line tools for processing point clouds. All tools can be run from the command line via the new native GUI *laslook* or via toolboxes (ArcGIS PRO, QGIS, FME, Erdas).
OBIA & LiDAR: Point Clouds and Object primitives in theory and practice
This session will provide an introduction to LiDAR technology, covering its fundamentals and the types of LiDAR platforms, such as airborne, terrestrial, and mobile systems, each tailored to specific applications in spatial data collection. It will then focus on the theoretical and practical aspects of using LiDAR data and Object-Based Image Analysis (OBIA) for object recognition and classification within spatial point clouds. Emphasis will be placed on segmentation and classification techniques that enable automated extraction of information, such as building outlines or tree canopy. Participants will explore tools and approaches for developing classification rules based on geometric and radiometric properties of data, as well as methods for assessing the transferability of developed models across different datasets.
Basics on SAR with a practical session on preprocessing of SAR image
Synthetic Aperture Radar (SAR) are active sensors that utilize electromagnetic waves in the microwave spectrum to analyze the geometric and dielectric properties of target objects. Thanks to their active nature and operating frequency, SAR systems are not dependent on sunlight and are nearly insensitive to weather conditions like cloud cover. This makes them invaluable for a wide range of applications. This lecture introduces the fundamental principles of SAR, including SAR interferometry, data processing techniques, and real-world applications. Additionally, participants will gain hands-on experience using open-source software and cloud-based platforms for satellite data analysis, enabling them to effectively process and interpret SAR imagery.