I am a Ph.D. student and a researcher in Leibniz University Hannover, Institute of Cartography and Geoinformatics. My research topic is application of deep learning and computer vision for detection and description of terrain structures related to historical mining and archaeology. I am focusing mainly on using Airborne Laser Scanning (ALS) data and its derivatives (e.g., Digital Terrain Models) to detect objects of interest for archaeologists.

I originally come from Ghazni, Afghanistan, but moved to Turkey in 2009 to study Computer Engineering in Middle East Technical University. Afterwards, I got a master’s degree in Artificial Intelligence from Polytechnic University of Catalonia in Barcelona, Spain. Since 2017, I am living in Hannover where I continue my Ph.D. studies and work as a researcher at Leibniz University.


  • Artificial Intelligence
  • Deep Learning
  • Computer Vision
  • Object Detection


  • Ph.D. in Geodesy and Geoinformatics, (Ongoing)

    Leibniz University Hannover

  • M.Sc. in Artificial Intelligence, 2017

    Polytechnic University of Catelonia

  • B.Sc. in Computer Engineering, 2015

    Middle East Technical University



Deep Learning Researcher

Institute of Cartography and Geoinformatics

Apr 2017 – Present Hannover

Quality Assurance

Typeform S.L.

Jul 2016 – Mar 2017 Barcelona

Java Developer

Open University of Catalonia

Feb 2016 – Jul 2016 Barcelona

Recent Posts

Image Classification With Tensorflow

This is an article showing the process of data preparation, creating a deep learning classification model, training, and evaluating it …

Building Blocks of Deep Neural Networks

This post describes basic building bloks of deep learning

Deep Learning Project Workflow

This post contains an abstract overview for a workflow in a deep learning project

Giving Back

This is my first blog post where I share implementations of some of the well known deep learning methods using Tensorflow with Keras.


Tensorflow Implementations of Deep Learning Models

This project is about Tensorflow implementations of deep learning architectures using Keras.

Recent Publications

Quickly discover relevant content by filtering publications.

Semantic Segmentation of Manmade Landscape Structures in Digital Terrain Models

We explore the use of semantic segmentation in Digital Terrain Models (DTMS) for detecting manmade landscape structures in …

Object Instance Segmentation in Digital Terrain Models

We use an object instance segmentation approach in deep learning to detect and outline objects in Digital Terrain Models (DTMs) derived …

Deep learning for archaeological object detection in airborne laser scanning data

It is important to preserve archaeological monuments as they play a key role in helping us understand human history and their …

Classification of laser scanning data using deep learning

In the last couple of years Deep Learning has gained popularity and shown potential in the field of classification. In contrast to 2D …