"Object Segmentation and Detection in Remote Sensing Images Based on Deep Learning Methods: Theory and Practice Using PyTorch Framework"

High resolution remote sensing images contain detailed information, making it possible to recognize objects within them. This interactive tutorial will begin by introducing some of the state-of-the-art detection and segmentation methods based on deep learning and explaining their building blocks. The focus will be on road, building, and vehicle classes as well as crowd detection and density estimation, which will be highlighted with visual demonstrations. This will be followed by a quick hands-on experiment involving the audience. During the tutorial, some networks will be processed live, and their training process as well as their results will be analyzed and discussed jointly with the audience. Concluding this interactive tutorial will be a live demonstration of real-time object detection on an embedded board.

The PyTorch framework will be used throughout the tutorial.




Seyed Majid Azimi

Seyed Majid Azimi is currently pursuing the Ph.D. degree with the Technical University of Munich (TUM), Munich, Germany, with a focus on traffic and infrastructure monitoring from remote sensing data using deep learning methods. Since 2016, he has been a Scientific Researcher with the Department of Photogrammetry and Image Analysis, Remote Sensing Technology Institute, German Aerospace Center (DLR), Weßling, German. His research interests include (embedded) computer vision and machine learning for object detection, segmentation, and tracking.


Reza Bahmanyar

Reza Bahmanyar is a research associate at the Photogrammetry and Image Analysis department of the German Aerospace Center (DLR), holding M.Sc. and Ph.D. degrees in Computer Sciences from Saarland University (Saarbrücken) in 2012 and Technical University of Munich in 2016, respectively. His main research interests include Machine Learning, Computer Vision, Image Processing, Data Mining, and Artificial Intelligence with the application in the Remote Sensing domain.




Corentin Henry

Corentin Henry is pursuing a Ph.D. degree at the Department of Photogrammetry and Image Analysis in the Remote Sensing Technology Institute of the German Aerospace Center (DLR-IMF). He received a M.Sc. degree in Computer Science from ISEN-Lille Engineering School, Lille, France, in 2017. His main research topic is Computer Vision and focuses on the application of Deep Learning to Remote Sensing tasks, especially to the extraction of roads topology in aerial imagery.


Workshop date: 15th Oct. 2019   8:30-12:30


"Terrestrial laser scanner and accelerometer based deformation monitoring of civil engineering infrastructures"

Today, short- and long-term deformation monitoring of civil engineering infrastructures based on terrestrial laser scanner (TLS) and accelerometers has received considerable attention. In the scope of this workshop, we will address three major issues in this regard. 

The first part will focus on theoretical aspects including deformation monitoring models (congruence and kinematic), sensor aspects (TLS, accelerometer), data acquisition and geo-referencing, approximation approaches (e.g. curve and surface based), and basics on time series analysis.  
In the second part, we are demonstrating the theoretical issues with practical examples as follows:

a) Introduction of the used sensors (accelerometer, TLS)

b) Data acquisition and measurement procedures

c) Analysis concepts (basic concepts in time series analysis as well as adjustment calculations) including surface computations and comparisons

d) The above steps are presented in case studies for either congruence or kinematic deformation analysis of static and kinematic objects.

The third part is dedicated to practice the time series analysis of datasets from the accelerometers and 2D profiles of the TLS. In addition, cloud-to-cloud comparison is presented and practiced based on a freely available software tool.





Jens-André Paffenholz

Dr.-Ing. Jens-André Paffenholz received his Dipl.-Ing. and Ph.D. in Geodesy and Geoinformatics at the Leibniz University Hannover in 2006 and 2012, respectively. Since 2014, he has been the leader of the working group Terrestrial Laser Scanner Based Multi-Sensor Systems | Engineering Geodesy at the Geodetic Institute of the Leibniz University Hannover. His research profile is based on laser scanning and multi-sensor systems with the aim of an efficient three-dimensional data acquisition for monitoring and change detection of natural and anthropogenic structures. He is active in national (DVW e. V.) and international scientific associations (working group chair of IAG WG 4.1.3).


Mohammad Omidalizarandi

Mohammad Omidalizarandi received a M.Sc. degree (Geomatics Engineering) from the University of Stuttgart (2011). Since 2014, he has been at the Geodetic Institute of the Leibniz University Hannover. He is currently pursuing a Ph.D. degree with a focus on vibration analysis of bridge structures using low-cost accelerometers and an image-assisted total station. His research areas are: sensor calibration and error modelling (i.e. terrestrial laser scanning, total station, digital camera and accelerometer), time series analysis, robust parameter estimation, adjustment computation, sensor integration, feature extraction from digital images and point clouds, vibration analysis and deformation monitoring.





Ingo Neumann

Prof. Dr.-Ing. Ingo Neumann received his Dipl.-Ing. and Ph.D. in Geodesy and Geoinformatics at the Leibniz Universität Hannover in 2005 and 2009, respectively. From 2009 to 2012 he was head of the Geodetic Laboratory at the University FAF Munich. Since 2012 he is Full Professor in the field of Engineering Geodesy and Geodetic Data Analysis at the Geodetic Institute of the Leibniz Universität Hannover. From 2014 on he is also head of the Geodetic Institute. His research areas are: adjustment theory and error models, multi-sensor systems, quality assessment, geodetic monitoring, terrestrial laser scanning, and automation of measurement processes. He is active in national and international scientific associations and member of the German and International Organization of Standardization (DIN and ISO).

Workshop date: 15th Oct. 2019   13:00-18:00

"Projections of Climate Change, Evapotranspiration, and Net Primary Production over Iran's Climatic Zones using General Circulation Models"

We will explain how to use climate-based models for projections of evapotranspiration (ET) and net primary production (NPP) over Iran's climatic zones under changing climate by General Circulation Models (GCMs). Additionally, the sensitivity of ET and NPP in response to climate change across diverse climates of Iran will be examined. This cooperative workshop is a part of a joint project supported by Iran National Science Foundation (INSF) and Chinese Academy of Sciences (CAS)  to enhance the scientific connections and cooperation between Belt and Road Initiative (B&R) countries.



TANG Qiuhong

TANG Qiuhong received the Ph.D. degree in Civil Engineering from the University of Tokyo, Tokyo, Japan, in 2008. He was a research associate at the department of civil and environmental Engineering, University of Washington from 2006-2010. He is currently a full Professor with the Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences. His current research interests include Remote sensing hydrology, Climate change, Hydrological modeling, Hydrological monitoring and forecast.


Pedram Attarod

Pedram Attarod received the Ph.D. degree in forest micrometeorology from the Tokyo University of Agriculture and Technology, Tokyo, Japan, in 2005. He is currently an associated professor with the Department of Forestry and Forest Economics, Faculty of Natural Resources, University of Tehran, Tehran, Iran. He has more than 40 papers mainly focused on forest eco- hydrology and impacts of climate change on water cycle in forests. He has an international joint-project with the Key Laboratory of Water Cycle and Re-lated Land Surface Processes, Institute of Geographic Sciences and Natural Research, Chinese Academy of Sciences, Beijing, China.