Home ITCAbout ITCCentres of expertiseCenter of Expertise in Big Geodata ScienceBig Geodata StoriesDeep Learning-based Change Detection and Classification for Airborne Laser Scanning Data
Waleed Alzuhair, flickr

Deep Learning-based Change Detection and Classification for Airborne Laser Scanning Data

Do you need a geospatial computing platform?

Urban change detection plays a critical role in many domains such as city planning, infrastructure development, risk assessment, and land-use planning. However, the accurate classification of different types of changes in a 3D urban environment remains a challenging task. Conventional methods, which typically involve transforming the data into a DSM or Voxels, often fall short in managing the complexity of point clouds. On the other hand, more complex deep learning models have shown promise but still face challenges in real-world applications. The Master's thesis aimed to address these challenges by implementing and evaluating three models: Random Forest, Fully Connected Neural Network, and Convolution Neural Network.


Figure 1. An overview of the methodology (left) and AHN acquisition methods (right)

The models are trained and validated on two distinct datasets, simulated Urb3DCD dataset and a real-world AHN dataset from the Netherlands, and are later assessed on their ability to accurately classify different types of changes. Findings reveal that while all models demonstrate strong performance, each has its strengths and weaknesses. The results obtained from the simulated dataset were outstanding. When applied to the AHN dataset, the performance is found to be highly sensitive to feature selection, the quality of training data, and how representative the data is. The direct comparison of classes between the simulated and real datasets indicates superior results from the simulated dataset, suggesting the need for better-quality training data for real-world applications.


Figure 2. Prediction on an AHN tile, a) Ground truth, b) RF, c) FCNN, and d) CNN. Blue colour represents non-change, yellow/ light green represents new buildings and red represents demolition.

The primary challenge in this project was managing the vast volume of point-cloud data. Unlike images, point clouds represent three-dimensional data, allowing for multiple data points at the same location, a scenario often encountered in forested areas. The sheer volume of this data exceeded the processing capabilities of my local PC.  Geospatial Computing Platform was instrumental in overcoming this hurdle by offering a robust computational environment that enabled me to test various models extensively until I achieved satisfactory results. The platform's user-friendly interface significantly enhanced my thesis work, making the research process not only more efficient but also more enjoyable. I am deeply thankful for the support Centre of Expertise in Big Geodata Science (CRIB) provided.


Figure 3. Visualization of the thesis product (up) and comparison of accuracies with different models (down).

In an era where data quality and volume are improving, the demand for platforms that enable researchers to test their ideas is increasing. The Geospatial Computing Platform meets this need effectively, facilitating research progress by eliminating computational delays and other non-essential obstacles. This support benefits not only the researchers but also advances the interests of the university and the broader scientific community.

For more information: 

Nofulla, Jorges (2023) Deep Learning-based Change Detection and Classification for Airborne Laser Scanning Data.

GitHub repository: Point-Cloud-Urban-Change-detection

Ir. Jorges Nofulla

I am a GIS specialist and a Data scientist, with a keen focus on leveraging machine learning, AI, and NLP to enhance our understanding and analysis of spatial and non-spatial datasets. My career began in geodesy engineering in Tirana, Albania, where I developed a solid foundation in geospatial technologies. Since then, I have expanded my expertise to include geo-information science, earning a Master's degree in the field, along with a Bachelor's in Geodesy Engineering. My current work involves exploring the potential of deep learning and AI in processing and interpreting complex data structures, aiming to contribute valuable insights to the field.

dr.ir. S.J. Oude Elberink (Sander)
Associate Professor

Sander Oude Elberink graduated as Geodetic Engineer from Delft University of Technology in 2000, and finished his PhD on the Acquisition of 3D Topography at ITC University of Twente, in March 2010. Since then he works as assistant and later associated professor at the Earth Observation Science department of ITC. He is teaching photogrammetry and laser scanning in various courses, he is course coordinator of the Joint Educational Program with IIRS Dehradun, India.

Dr. habil. Michael Ying Yang

Michael Yang is currently employed as full professor at the University of Bath, UK. He has been working at ITC from 2016-2024 in the Earth Observation Science department at ITC. His main academic interests are in the overlap of Computer Vision and Photogrammetry.