Seyed Ehsan Khankeshizadeh - support disaster assessment - Best Researcher Award
Department of Photogrammetry and Remote Sensing - Iran
AUTHOR PROFILE
💻 EXPERTISE IN PYTHON PROGRAMMING AND DEEP LEARNING
Seyed Ehsan Khankeshizadeh is a skilled Python programmer with a profound interest in Deep Learning techniques. He specializes in semantic segmentation using advanced UNet-based structures such as UNet, ResUNet, RecurrentResidualUNet, UNet++, and Attention-U-Net. His work primarily focuses on building detection, damage assessment, and forest change detection through the processing of satellite sensor imagery.
🛰️ SATELLITE IMAGE PROCESSING AND ANALYSIS
Ehsan has extensive experience in satellite image processing, particularly in ENVI. He excels in image pre-processing techniques including radiometric and geometric calibration, image registration, Pan Sharpening, Masking, and Mosaicking. His expertise extends to image classification using both supervised and unsupervised methods, with a strong focus on post-classification processes.
🌍 CHANGE DETECTION AND REMOTE SENSING
A significant aspect of Ehsan’s research involves change detection in remote sensing data. He applies Synthetic Aperture Radar (SAR) image processing and utilizes various remote sensing spectral indices within platforms like ENVI and Google Earth Engine to analyze environmental changes and their impact.
🧠 PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE
Ehsan integrates Machine Learning and Artificial Intelligence techniques into his work, leveraging the capabilities of ENVI, SNAP, and Python. His focus on pattern recognition enables him to develop sophisticated models for analyzing complex satellite data, contributing to advancements in geospatial intelligence.
🌳 3D POINT CLOUD PROCESSING AND FOREST CHANGE DETECTION
Ehsan’s expertise also includes 3D point cloud processing, using tools like LAStools and Cloud Compare to analyze spatial data. His work on forest change detection using recurrent residual-based U-Net models demonstrates his commitment to preserving natural resources and understanding environmental dynamics.
📚 PUBLISHED RESEARCH IN SATELLITE IMAGERY ANALYSIS
Ehsan is an accomplished researcher with several publications to his name. His work on FCD-R2U-net for forest change detection and the development of the Weighted Ensemble Transferred U-Net Model (WETUM) for post-earthquake building damage assessment showcases his contribution to the field of remote sensing and geospatial analysis.
🏆 INNOVATOR IN BUILDING DETECTION AND DAMAGE ASSESSMENT
Ehsan's innovative approach to building detection using Dual Attention Residual-based U-Net (DAttResU-Net) highlights his ability to push the boundaries of satellite imagery analysis. His models are instrumental in generating building change maps, aiding in disaster response and urban planning efforts.
NOTABLE PUBLICATION
A Novel Weighted Ensemble Transferred U-Net Based Model (WETUM) for Postearthquake Building Damage Assessment from UAV Data: A Comparison of Deep Learning-and Machine Learning-Based Approaches
Authors: E. Khankeshizadeh, A. Mohammadzadeh, H. Arefi, H. Li, J. Li
Journal: IEEE Transactions on Geoscience and Remote Sensing
Year: 2024
FCD-R2U-net: Forest change detection in bi-temporal satellite images using the recurrent residual-based U-net
Authors: E. Khankeshizadeh, A. Mohammadzadeh, A. Moghimi, A. Mohsenifar
Journal: Earth Science Informatics
Year: 2022