Geomatics Engineering

Graduation Project Topics

Hacettepe University GMT403-GMT404 2021-2022

TABLE OF CONTENTS

Project Titles Page

An approach for burned area extraction from Sentinel-2 and Landsat-8 images 1
An approach for tree detection using airborne LiDAR data 2
Design and development of a CCD camera-assisted theodolite system 3
Building architecture independent data module for geopotential functionals derived from spherical harmonics 4
Design and Implementation of a Point Cloud-based Powerline Corridor Mapping System 5
Design and Implementation of an Optical-Thermal Data Integration Methodology to Investigate Heat Losses from Building
Windows 6
Urban Design Concept: Design of walkability and cycling paths with GIS and suitable multi-criteria decision-making (MCDM)
methods 7
Development of an Indoor Positioning System for Shopping Malls 8
Design of a building damage assessment tool 9
Development of an Android App for Underground Navigation and Mapping 10
Modeling of the land surface temperature on land cover pattern 11
Design of indoor parked location finding 12
Design of an Artificial Intelligence guided geodetic time series predictor 13
Development of user-friendly geodetic surveying application 14
Simulating the observations of the VLBI sessions 15
Optimization of the observation model for the VLBI Intensive session analysis 16
Public Transportation Modeling with a Web and Mobile Interface 17
Design and Development of TUCBS Compliant Beytepe Campus Pedestrian Navigation System 18
Development of an Indoor Positioning and Navigation System based on 360-degree cameras 19
Development of a 3D Satellite Orbit Simulator Software 20
Development of a QGIS plugin for time-series analyses and forecasting of the COVID-19 cases 21
Development of a WEB Based Geodetic Toolbox that solve a variety of problems in Geodesy 22
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title An approach for burned area extraction from Sentinel-2 and Landsat-8 images

Supervisor Prof. Dr. Mustafa Türker

Expected team size 2 students
Problem Definition Forest fire is one of the most devastating natural events that most modifies the ecosystems and plays an
important ecological role in most of the Earth’s surface. It is estimated that wildfires yearly affect
about 50 million hectares all over the Earth’s surface. Recently, forest fires have also been quite
destructive in Turkey, where approximately 150,000 hectares were destroyed and thousands of people
were evacuated due to long and vast fires, in southwestern Turkey.

Accurate and timely extraction and quantification of the burnt areas are quite necessary to assess the
size and amount of the damaged areas, address the post-fire management activities, and implement
short- and long-term landscape strategies. In this respect, space-based remote sensing provides
systematic timely images to monitor both man made or natural changes that occur in the Earth’s
surface. In particular, monitoring active fires and detecting the burnt areas are achievable with the use
of satellite images due to the spectral, thermal and radar backscatter changes induced by fires. The
freely available timely Sentinel-2 and Landsat-8 satellite images are very useful for post-fire burnt area
detection.

Objectives and constraints The main objective of this study is to develop an approach for the automatic extraction of burned areas
from the integrated use of Sentinel-2 and Landsat-8 images.

• The approach should be applicable for Sentinel-2 and Landsat-8 images.

• The approach should extract the burned areas correctly with low error values.
• The approach should operate on those image data sets with the spectral and spatial
characteristics similar to Sentinel-2 and Landsat-8 satellites.
• The images to be used should be cloud free images.
• For the validation of the results, accurate and reliable reference data set is required.

Milestones and • Performing literature survey by the end of October 2021.

Deliverables • Selecting study areas and image data sets by the end of October 2021.
• Preparing reference data for use in validation of the results by the end of November 2021.
• Designing the conceptual framework of the proposed approach by the end of December 2021.
• Developing a program using Matlab or an open source software to implement the approach by
the end of February 2022.
• Performing data analysis on study areas using the developed program by the end of March
2022.
• Writing the project report by the end of April 2022.

1
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title An approach for tree detection using airborne LiDAR data

Supervisor Prof. Dr. Mustafa Türker

Expected team size 2 students
Problem Definition Accurate detection and characterization of the distribution of individual trees have significant
implications in sustainable forest management and urban applications. Once detected, several structural
attributes, such as number, height, crown diameter, basal area, species type, density, and wood volume
can be estimated.

Traditional methods, such as field inventory and aerial photo interpretation are labor-intensive and
time-consuming. Early methods of tree detection were mainly based on optical imagery. The methods
based on optical imagery have limitations, such as they could not acquire the accurate 3D information
of trees, optical images are not able to penetrate the tree canopy to measure tree structure. These
limitations can be overcome using lidar (light detection and ranging) technology.

Airborne lidar is an active remote sensing technology that provides very accurate distance
measurements between the platform and the surface. In the last decade, lidar has become a promising
technique and has been frequently and successfully used in detecting trees in forested and urban areas.
When compared with optical imagery, lidar data provides substantially more information with regards
the vertical structure of the trees. In the last two decades, a number of tree detection algorithms have
been proposed. In this project, an approach will be developed to detect individual trees from discrete
return airborne lidar point clouds. The methods will extract individual trees directly from the lidar
point clouds. The experiments will be conducted in both rural and urban areas. The results will be
evaluated using precision, recall, and F-score.

Objectives and constraints The main objective of this study is to develop an approach to detect individual trees directly from
discrete return airborne lidar point clouds.

• The approach should perform well in detecting trees directly from discrete return airborne
lidar point clouds.
• The approach should work on lidar data sets that contain both ground and above ground
points.
• The approach should have good potential for use in typical urban or forested areas.
• The approach should provide the numbers and locations of the detected trees.
• The approach will require lidar point clouds with medium and higher point densities.

Milestones and • Performing literature survey by the end of October 2021.

Deliverables • Selecting study areas and the corresponding lidar data sets by the end of October 2021.
• Preparing reference data for use in validation of the results by the end of November 2021.
• Designing the conceptual framework of the proposed approach by the end of December 2021.
• Developing a program using Matlab or an open source software to implement the approach by
the end of February 2022.
• Performing data analysis on study areas using the developed program by the end of March
2022.
• Writing the project report by the end of April 2022.

2
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Design and development of a CCD camera-assisted theodolite system

Supervisor Prof. Dr. Aydın Üstün

Expected team size 3 students
Problem Definition In order to measurement horizontal and vertical angles at observation point, theodolites are utilized for
classical optical observations. In some cases, direction observations are required for a large number of
objects within telescope's view field, rather than optical direction readings for a particular target. These
kinds of objects might be identified from an image by a camera system like CCD (Charge Coupled
Device) which is mounted to telescope of the theodolite. The CCD assisted theodolite helps to
monitoring of deformation area, such as mining field, dam or engineering buildings.

The CCD theodolite system should be considered and operated for terrestrial engineering, and
astrogeodetic measurements that mean zenith orientation of telescope. The mechanical interface of
CCD camera must be adaptable to the telescope of a geodetic instrument that enables high accuracy
angle measurements (better than 1 arc-second).

Objectives and constraints It is aimed to develop and implement a theodolite system that can make high-accuracy directional
observation for video imaged objects by CCD camera unit.

• The system should be operated for zenith directed measurements.

• Preferred CCD camera with firewire cable and related accessories should not decrease
accuracy of optical direction measurements.
• Focusing should be controlled by an integrated computer to the CCD camera.
• An image processing software should be capable for identifying objects to be measured.

Milestones and
Deliverables • Reporting an analyze for market products of CCD camera devices by the end of October
2021.
• Designing mechanical interface of the appropriate CCD unit for the replacement of eyepiece
of telescope with the CCD unit by the end of November 2021.
• Providing a technical guidance for imaging system associated with image processing software
by the end of semester.
• Implementation of CCD assisted theodolite system and calibration for imaging and object
identification software by the first week of April 2022

3
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Building architecture independent data module for geopotential functionals derived from
spherical harmonics

Supervisor Prof. Dr. Aydın Üstün

Expected team size 3 students
Problem Definition Geopotential is a physical quantity that describes the gravity of any astronomical object, such as the
earth, moon, sun, or galaxy. Within the earth's gravitational field, geopotential is measured as the
potential energy of a unit weight mass object. If the distribution and density of the earth masses are
known very well the geopotential and its derivatives can be analytically represented and numerically
computed by means of spherical harmonics. Global Geopotential Model (GGM) used to calculate the
geopotential consists of a series of constant numerical harmonic coefficients truncated at a specific
degree (and order) to the model expansion. Point or grid wise geopotential data sets are required to
determine heights or to map the earth's gravity field in geodetic sciences and related disciplines. In this
project, a geopotential data module producing NetCDF data set, which is a gridded scientific data in
binary format, is wanted to build in an architecture independent environment.
Objectives and constraints It is aimed to develop and implement a program module that creates NetCDF format gridded data set
for geopotential quantities from earth gravity model (or GGM).

• The program module will be written in C programming language.

• The geopotential module must be integrated into the previously developed C APIs of the
General Mapping Tools (GMT) for the NetCDF grid structure.
• The module must be invoked by command line driven arguments, inputs and options.
• Spherical harmonic models are released by the International Center of Earth Gravity Models
(ICGEM). The module should be able to read ICGEM ascii format.

Milestones and
Deliverables • Understanding NetCDF format and reporting GMT C API implementation for simple gridded
data examples (GMT403).
• Performing practical applications spherical harmonics in Matlab and reporting theory and
practice to the computation of geopotential quantities for Earth Gravity Models (GMT403).
• Designing program module of gridded geopotentials based on GMT Grid API tools
(GMT403).
• Developing C program module that calculates geopotantial functionals and creates grid
structure for output which is platform independent (GMT404)
• Performing program tests and map examples for numerical examples in both global and
regional scales (GMT404)
• Writing thesis and presentation (GMT404)

4
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Design and Implementation of a Point Cloud-based Powerline Corridor Mapping System

Supervisor Assoc. Prof. Dr. Ali Özgün OK

Expected team size 3 students
Problem Definition Inspection of power lines is a critical task for power distribution companies. To avoid an unintentional
service interruption, it is necessary to schedule the removal of obstacles close to the powerline. To avoid
intrusion inside the powerline corridor, it is especially important to keep vegetation from growing by
pruning or cutting it. A potential obstacle can be defined as any object within the transmission right of
way or powerline corridor. Because these inspections are costly to conduct on the ground, remote sensing
based on airborne light detection and ranging (LiDAR) data is becoming more popular. We can directly
collect a 3-D point cloud of the observed scene using LiDAR technology, which can then be used to
identify the obstacles inside the powerline corridor after classification.

Objectives and constraints In this graduation project, the main tasks are (i) automatic extraction of the wires and (ii) identification
of obstacles close to the powerlines.

• The extraction of the wires should be carried out using LiDAR first returns solely.
• The extraction principle should rely on a supervised classification procedure, and should integrate a
deep learning framework. The expected overall automated classification performance should be at
least 85% for the test set.
• For every detected wire, a buffer area has to be defined; this buffer area has a cylinder shape (tube)
with the longitudinal axis along the wire and radius defined by the safety requirements.
• A LiDAR point should be labelled as obstacle if it does not exceed a specific distance (i.e. 5m) from
the wire or if the point is inside a defined corridor.
Milestones and 2021-2022 Fall Semester (GMT403):
Deliverables • An analysis on the design alternatives for the supervised classification procedures that can be used
for point clouds should be carried out.
• The architecture that would be used to develop a powerline corridor mapping system should be
proposed.
• An evaluation considering the availability of datasets, training/test scenarios, and related
hardware/software requirements should be provided.
• A (detailed) report covering the analyses carried out and the architecture proposed for the given task
should be submitted.
2021-2022 Spring Semester (GMT404):
• The verification of the proposed architecture should be completed by a plugin implemented in
commercial/open source GIS software.
• A final project report covering all above aspects should be submitted.
2021-2022 Summer Semester (if desired by the students and deemed appropriate by the supervisor):
• Submission of a conference contribution related to the graduation project topic.

5
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Design and Implementation of an Optical-Thermal Data Integration Methodology to Investigate
Heat Losses from Building Windows

Supervisor Assoc. Prof. Dr. Ali Özgün OK

Expected team size 3-5 students
Problem Definition Buildings consume a significant portion of the total energy consumed by humans. Building energy
efficiency should be improved as a result of climate change, rising energy costs, and energy performance
directives. Buildings with high energy efficiency can now be built thanks to new technologies, but it is
still critical to inspect specific elements of (old) buildings, i.e. windows, for potential heat losses. A
thermal infrared (IR) camera can be used to inspect buildings remotely, and optical images can be used
to identify elements of buildings. In order to combine IR images with high resolution optical images for
extracting building heat loss information, the data must be spatially referenced. The assignment of all
information to a 3D building model should be done automatically and can be achieved via texture
mapping.

Objectives and constraints In this graduation project, the main tasks are (i) optical-thermal image registration and (ii) specific
identification of heat losses from building windows.

• The optical-thermal image registration should be carried out using one of the feature matching
strategies. The expected overall feature matching performance should be at least 75% for similar
viewing conditions (< 10°).
• The extraction of building windows should be carried out using optical images solely. The extraction
principle should rely on a supervised classification procedure, and should integrate a deep learning
framework. The expected overall automated classification performance should be at least 85% for
the test set.
• All information extracted should be mapped to a 3D building model and should be visualized via
texture mapping.
Milestones and 2021-2022 Fall Semester (GMT403):
Deliverables • An analysis on the design alternatives for the optical-thermal image registration and supervised
classification procedures that can be used for the extraction of building windows should be carried
out. The requirements will also be determined.
• The architecture that would be used to investigate heat losses from building windows should be
proposed.
• An evaluation considering the availability of datasets, training/test scenarios, and related
hardware/software requirements should be provided.
• A (detailed) report covering the analyses carried out and the architecture proposed for the given task
should be submitted.
2021-2022 Spring Semester (GMT404):
• The verification of the proposed architecture should be completed by a plugin implemented in
commercial/open source software.
• A final project report covering all above aspects should be submitted.
2021-2022 Summer Semester (if desired by the students and deemed appropriate by the supervisor):
• Submission of a conference contribution related to the graduation project topic.

6
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Urban Design Concept: Design of walkability and cycling paths with GIS and suitable multi-
criteria decision-making (MCDM) methods

Supervisor Assoc. Prof. Dr. Cevdet Coşkun AYDIN

Expected team size 2 students
Problem Definition The development of active transportation vehicles has brought some urban problems. Efforts are being
made to solve many problems such as air pollution, traffic, health, greenhouse gas emissions. The most
effective ways to solve these problems are walking, cycling, and increasing the use of electric vehicles.
The effective implementation of these solutions depends on proper planning in urban areas. The use of
multi criteria decision making (MCDM) methods are very important and common in making appropriate
choices according to objective criteria in decision-making processes.

Objectives and It is aimed to increase the livability level of urban areas and to create sustainable urban models by
constraints using AHP, TOPSIS and WLC methods in the planning of walkability and bicycle paths.

• Data on settlement areas, urban development areas, roads, transportation networks and
important places in cities should be provided.
• Sub-criteria related to main criteria should be determined.
• Determining the weights of the criteria with Analytic Hierarchy Process (AHP).
• Transferring the criteria to the GIS environment in a common coordinate system.
• Classification of layers according to the values of the sub-criteria.
• Determination of the study area.
• Weighting the layers and determining the most suitable areas with the WLC method.
• Determining the priority order of the most suitable places with the TOPSIS method.
Milestones and
Deliverables • Complete the literature review by the end of October 2021.
• Determining the study area and providing the necessary data by mid-December 2021.
• Making the implementation stage and creating the first reports/applications until the end of
January 2021.
• Completion of all necessary study by April and delivery of the work.

7
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Development of an Indoor Positioning System for Shopping Malls
Supervisor Assoc. Prof. Dr. Metin Nohutcu
Expected team size 3 students
Problem Definition While GNSSs provide satisfactory positioning accuracy for outdoor environments globally, they
cannot be used inside buildings efficiently due to non-line of sight issues. On the other hand,
positioning and navigation necessities for indoor environments are growing each day with the
increasing number of complex buildings with high number of visitors/users, e.g. airports, train stations,
hospitals, museums, factories, shopping malls etc. Mobile phones, by far, are the mostly used devices
on the user-side for this purpose as they do not require additional costs. Therefore, the technologies
that the indoor positioning systems (IPSs) rely on are generally dependent on the built-in sensors
available in smart phones. There are a wide range of methods used in IPSs with their
advantages/disadvantages such as: wireless technologies, e.g. Wi‑Fi, Radio Frequency Identification
(RFID) visible light, Bluetooth, ultra‑wide band (UWB); computer vision-based systems employing
3D cameras or inbuilt smartphone cameras; inertial measurement sensors such as accelerometers,
gyroscopes, and magnetometers. A successful IPS should consider user needs besides the structural
limitations of the building that the system will be deployed in.

Objectives and constraints The aim of this project is to develop an IPS dedicated to shopping malls considering the user
expectations including customers, retailers and facility management.

• The expected positional accuracy should be within 3 m with 90% confidence level. The system
may rely on the measurements of a single sensor or multi-sensor fusion approach to achieve
desired accuracy.
• The system should include an end-user application that will run in mobile phones/tablets and a
software that will be managed by the administrators to follow the location of all users. Due to the
limitations of iOS to access some of the sensor measurements, the mobile app is expected to run
in Android devices.
• A prototype of the system should be installed for a demonstration. A limited budget will be
provided to the project team for the possible hardware expenses of the prototype.

Milestones and • Literature survey, reporting an analysis for the available IPS technologies/methods and decision
Deliverables for the selection of the methodology that will be used for the system by the end of October 2021.
• Determination of the services that would be provided to the users with the end-user application
and administrative software by the end of November 2021.
• Purchase of system hardware (e.g. beacons) that will be used for the prototype of the system
software by the end of December 2021.
• Completion of the software components with the essential services by the third week of March
2022.
• Implementation and demonstration of the prototype system by the first week of April 2022.

8
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Design of a building damage assessment tool

Supervisor Assoc. Prof. Dr. Saygin ABDIKAN

Expected team size 3 students
Problem Definition Natural hazards can cause loss of life and properties. In order to reduce the risk of collapses that may
occur after the event the level of the hazard of the objects at the ground should be identified.
Considering the big cities this might be time consuming. This kind of classification (i.e destroyed,
highly damages, moderately damaged and stable) can be identified rapidly using remote sensing
images which covers large areas.

The tool should be considered for estimating the building's hazard level after a devastating event such
as an earthquake or flood. The output product which is going to show damage grade of the objects
should have high accuracy in order to provide reliable geospatial information for decision makers.

Objectives and constraints It is aimed to develop an assessment tool that can detect damaged buildings accurately.

• The tool should detect buildings after a hazard

• High resolution satellite images should be provided
• The accuracy of the output should be checked

Milestones and • Reporting an analyze for the damage assessment approaches by the end of October 2021
Deliverables • Designing the flowchart of the tool by the end of December 2021
• Providing an image processing tool by the end of semester
• Implementation and production of a damage map by the end of April 2022

9
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Development of an Android App for Underground Navigation and Mapping
Supervisor Assoc. Prof. Dr. Metin Nohutcu
Expected team size 2 students
Problem Definition Navigation in outdoor environments has been improved in terms of both accuracy and reliability
especially after GNSSs. There have been several technologies, e.g. Wi‑Fi, Radio Frequency
Identification (RFID) visible light, Bluetooth etc., aided Indoor Positioning Systems (IPSs) for
positioning purposes in man-made structures where GNSS signals are not available also. However,
navigation in subterranean environments, such as mines, caves, or tunnels where no infrastructure is
available, is still a challenging problem to be enhanced.

Inertial sensors, i.e. accelerometers, gyroscopes, or magnetometers, may be the first option that spring
to mind for underground navigation as they can be used internally without depending on external
information for position calculation. However, inertial navigation relies on the previous information,
which results in integration drifts, i.e. the positioning error increases with time. Therefore, inertial
navigation systems (INSs) are frequently coupled with external data, e.g. GNSS measurements.
Although there are relatively higher accuracy inertial navigation devices, they are pretty expensive.

Objectives and The aim of this project is to develop an inexpensive solution for navigation and mapping in
constraints subterranean environments.

• Taking the cost into consideration, an Android platform (mobile phone or tablet) will be used in
the project.
• Additional inexpensive and lightweight equipment, such as laser distance meters or markers, can
be employed.
• The app to be developed should include functions for mapping purposes such as adding
benchmarks with timestamps besides position and orientation information.
• The system should be accurate enough to ensure the user to quit the environment safely.

Milestones and • Literature survey, reporting an analysis for up-to-date methods/technology by the end of October
Deliverables 2021.
• Design of the system, including calibration/correction methodology by the end of November
2021.
• Completion of the app with the essential functions by the third week of March 2022.
• Implementation and demonstration by the first week of April 2022.

10
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Modeling of the land surface temperature on land cover pattern

Supervisor Assoc. Prof. Dr. Saygin ABDIKAN

Expected team size 3 students
Problem Definition Urban planning includes several parameters for a sustainable city life. One of the important parameters
is the balance between urban expansion and thermal environment. However, anthropogenic activities
such as ports, thermal power plants, and other urban structures cause increase of temperature which
reduce life quality and health problems.

Remote sensing images can be used to determine land surface temperature (LST) which can be
correlated with the distribution of the land cover types such as urban, water and green space. The LST
maps should be also considered for engineering structures such as highways, bridges and dams. For the
analysis, land cover maps might be identified through multi-spectral images that should have high
accuracies (>85%).

Objectives and constraints It is aimed to develop and implement a methodology that can determine temporal changes of LST and
its relationship with land surface objects.

• The methodology should be operated for LST estimations

• The approach should be applicable for thermal and multi-spectral bands
• Geospatial relationship should identify different surface objects.
• An image processing software should be capable for identifying objects to be measured.

Milestones and
Deliverables • Reporting an analyze for the LST extraction approaches by the end of October 2021
• Designing the flowchart of the tool by the end of December 2021
• Providing an image processing design by the end of semester
• Extraction of land cover types by the end of March 2022
• Implementation and production of a damage map by the end of April 2022

11
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Design of indoor parked location finding

Supervisor Assistant Prof. Gonca Okay Ahi

Expected team size 3 students
Problem Definition Generally, in large indoor areas (e.g. shopping malls etc.) people have tendency to forget to where they
have parked. Sometimes the search of the corresponding location may take times especially in some
countries where huge designs of center of attractions are available and it is important if an emergency
situation occurs. In this case, it would be practical to develop a tool which will guide you to the
corresponding location easily and quickly independent from the help of the GPS system which does
not work in such areas.

Objectives and constraints It is aimed to develop and implement a mobile phone application which will use available
accelerometer/gyroscope sensor in the smartphones to find the parked location.

• The system should operate at indoor areas.

• The design should take care about the sensor sensibility of the common smartphones.

Milestones and
Deliverables • Reporting an analyze for the available sensors in smartphones, their uses and the sensor
sensibilities by the end of October 2021.
• Designing the raw sensor data extraction from smartphones by the end of November 2021.
• Implementing an artificial intelligence method to develop the strategy between physical
activity recognition and location information at the end of semester.
• Implementation of a dedicated mobile phone application by the first week of April 2022.

12
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Design of an Artificial Intelligence guided geodetic time series predictor

Supervisor Assistant Prof. Gonca Okay Ahi

Expected team size 3 students
Problem Definition In the age of technology, with the development of measuring devices, spatial data is available at huge
amounts which give birth to the concept of “big data”. The study of geodetic data as a big data may give
the opportunity of studying many geoscientific problems such as climate change, earthquake early
warning, sea level changes etc. To do so, the corresponding geodetic data has to be unified from different
sources and should be predicted with the help of artificial intelligence methods to provide the
preparedness for an emergency situation and the capability of monitoring natural resources.

Objectives and constraints Due to the above-mentioned scientific motivation, in this study, it is aimed to develop and implement
(1) a software which will use geodetic data as input and within the scope of the interested problem
definition (e.g. climate change), it will provide a parameter estimation (for a short or long-term) as an
output. At the end (2) a user-friendly mobile phone application should be produced.

• The spatial and temporal resolutions of the various geodetics data should be considered.
• The data amount is an important parameter which affect the intended short term or long term
parameter estimation. Besides, the performance of the use of artificial intelligence methods.

Milestones and
Deliverables • Deciding the problem definition and understanding/analyzing the associated geodetic data by
the end of October 2021.
• The preparation of the whole geodetic data and understanding adapted parameters estimations
methods by the end of November 2021.
• Implementing an artificial intelligence algorithm to provide a long-term and short-term
prediction at the end of semester.
• Testing the performances of the prediction method and implementation of a dedicated mobile
phone application by the first week of April 2022.

13
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Development of user-friendly geodetic surveying application

Supervisor Assistant Prof. Gonca Okay Ahi

Expected team size 2 students
Problem Definition The surveyor in the field is mostly focused with the proper data acquisition within the limited time for
an efficient time and cost management. Thus, the interpretation part is dedicated mostly to the office
hours. However, sometimes the acquired data may show some unwanted results which are not recognized
in the field. Or sometimes it appears that it might be interesting to lead the survey in a different way
according to the measured results. Thus, the surveyor return back to the field either to correct some
measurements or to acquire more data. In order to prevent these kind of problems and to manage better
the survey part, a new user-friendly strategy has to be considered.

Objectives and constraints For an efficient guidance to the surveyor, in this study, the development of a user-friendly geodetic
application is considered. With the help of location information, this application will be capable of
calculating every possible geodetic order of magnitude related with the gravity field (gravity anomaly,
geoid undulation, deflection of vertical etc.) within the studied area according to a predefined list of
available global geopotential models (GGM).

• The spatial resolution of the GGM’s are limited. They have to be considered as just
approximations of the real values of the interested order of magnitudes.
• There might be the problem of visibility of GPS satellites in the field so location information
should be defined in another way.

Milestones and
Deliverables • Understanding gravity field functional at the end of October 2021.
• Developing the corresponding codes which calculates all possible order of magnitudes by the
end of November 2021.
• Retrieval of location information from smartphones and integrating it in the related codes to
calculate the interested order of magnitudes and starting the development of the smartphone
application at the end of semester.
• Design of the smartphone application and testing the performance of the measured and
calculated results by the first week of April 2022.

14
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION

Project Title Simulating the observations of the VLBI sessions

Supervisor Dr. Kamil Teke

Expected team size 3 students
Problem Definition Construction, back-end equipment, and operating of a VLBI station cost too much. Thus, to maximize
the accuracy of the interested geodetic parameters, the observations of a VLBI network should be
simulated. Baseline length repeatability is one of the most reliable (unbiased) accuracy criteria (metric)
for the space geodetic techniques due to its independence from TRF and CRF datum definitions. The
scalar accuracy objective function of the design might be selected as minimizing the baseline length
repeatabilities. Through simulating the VLBI observations, the preceding objective function would be
ensured.

Objectives and constraints § This study aims to simulate the VLBI observations using several functional and stochastic models
to ensure the objective function of the design i.e. minimizing the baseline length repeatabilities.
§ Stochastic behaviors of the troposphere media and clock errors differs from one station to another.

Milestones and § Literature research by the end of October 2021.

Deliverables § Calculating the geometrical delays from a priori functional models, calculating the troposphere
delays from a turbulence simulator, calculating the clock errors from an integrated random walk
process, and finally deriving the simulated delay (observation) through summing up the
geometrical delay, troposphere delay, and clock error by the end of Fall semester of 2021-2022.
§ Analyzing the sessions of which observations are simulated, estimating the coordinates of and the
baselines between the VLBI stations, and calculating the baseline length repeatabilities by the end
of the Spring semester of 2021-2022.

15
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Optimization of the observation model for the VLBI Intensive session analysis
Supervisor Dr. Kamil Teke
Expected team size 3 students
Problem Definition VLBI Intensive sessions are carried out to monitor the universal time (UT1), the most variable Earth
orientation parameter, which is essential for the transformation between Earth and space fixed
coordinate systems. Among others, Intensive sessions are scheduled and organized regularly daily by
the International VLBI Service for Geodesy and Astrometry (IVS). Due to the financial constraints,
VLBI Intensive sessions are carried out with 2 to 4 VLBI radio telescopes observing for 1 to 2 hours.
This results in a few observations with bad sky coverage. Finally, the UT1 accuracy from Intensives is
worse than those derived from the 24 hour sessions. In addition, most of the parameters in the
observation model should be fixed to their a priori values e.g. TRF, CRF, polar motion, and nutation
due to the small degrees of freedom. This ends up with the propagation of the errors of the a priori
fixed parameters to UT1 in least-squares adjustment.
Objectives and constraints § This study aims to optimize the observation model for the VLBI Intensive session analysis. The
accuracy objective function can be selected as e.g. the formal errors of the UT1 estimates from the
Intensive sessions should be minimum and/or the RMS of the UT1 differences between Intensives
and 24 hour sessions should be minimum.
§ Basic constraints for this optimization problem are the number and sky coverage of the
observations, the troposphere delay modeling, the geometry of the baseline/s w.r.t. TRF.
Milestones and § Literature research by the end of October 2021.
Deliverables § Analysis of the Intensive sessions observed in 2021 with different observation models, a priori
models, and model parameters by the end of Fall Semester.
§ Developing computer functions for the statistical interpretation of the estimated UT1 time series
along with their formal errors.
§ Suggesting an optimal observation model and model parameters that ensures the objective
function by the end of the Spring Semester.

16
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Public Transportation Modeling with a Web and Mobile Interface

Supervisor Asst. Prof. Dr. Berk Anbaroğlu

Expected team size 3 students
Problem Definition Turkish cities have various road transportation mediums including, dolmuş or priate/public bus.
Although, these mediums are commonly used by most of the citizens, there is still a need for a web-
based system that shows the paths of these mediums in the major cities of Turkey including Ankara.
The lack of such a system is a real-life problem for tourists or students moving to a new city. This
project aims to design a participatory web-GIS that can collect and visualize data relevant to public
road transportation mediums. The collected data would also be made available through
OpenGeospatialConsortium (OGC) services such as WMS and WFS, and also an Application
Development Interface (API) would be developed to query the database.

Objectives and constraints The objectives of this project are:

• Designing a spatio-temporal database facilitated to the analysis of public transportation trips,

• Developing a web-based system that can
o obtain the path information of one or multiple paths from citizens,
o visualize the available paths,
o serve the collected data through open OGC standards and an API interface
• Developing an Android application that eases data collection.

Milestones and
Deliverables • Report and analysis of the available public transportation related systems in Turkey until the
end of October 2021.
• Designing the web and mobile interface for data collection until the end of November 2021.
• Implementation of the participatory web-based GIS and the mobile application by the first
week of April 2022.

17
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Design and Development of TUCBS Compliant Beytepe Campus Pedestrian Navigation System

Supervisor Asst. Prof. Dr. Berk Anbaroğlu

Expected team size 3 students
Problem Definition The Beytepe Campus of Hacettepe University has a multitude of buildings, which makes it difficult for
pedestrian navigation. There are some passages (e.g. next to the Biology department) connecting
different regions of the campus, yet, some students or visitors may not be aware of. In addition, there
are many buildings and lecture halls, where teaching and research are being conducted. The aim of this
project is to design and develop a web-based GIS that is easy to use for pedestrian navigation as well
as for infrastructure management. The data collected would be INSPIRE compatible so that it could be
distributed through the internet by a common standard. In this way, the data collected could also be
served to the Turkish National GIS (TUCBS). The OGC services including WMS and WFS would be
utilized, and the effectiveness of the system would be tested.

Objectives and constraints The objectives of this project are:

• Designing a database that can store roads and pavements as linear features, buildings as
polygon features and lecture halls as points. The constructed model would be topologically
correct and can be used for navigation.
• Developing a web-based system that can
o Find the shortest path between an origin and a destination,
o Visualize the shortest path,
o serve the collected data through open OGC standards and an API interface

Milestones and
Deliverables • Report and analysis of the available campus navigation systems until the end of October 2021.
• Collecting the data until the end of November 2021,
• Designing the web interface for data visualization until the end of December 2021,
• Implementation of the web-based GIS by the first week of April 2022.

18
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Development of an Indoor Positioning and Navigation System based on 360-degree cameras

Supervisor Asst. Prof. Dr. Murat Durmaz

Expected team size 3 students
Problem Definition Recent advancements in technology (communication, computation power and advanced vision systems)
and robotics results in increasing automation in the industry. When combined with sensor technologies,
the next wave of industrial revolution will be the humanless-factories, commonly known as lights-out
manufacturing. In such an environment, the intermediate products have to be transferred autonomously
in the factory through different processes. Such an autonomous navigation requires cheap, reliable, high
precision, fault tolerant navigation aid for Autonomous Guided Vehicles (AGV).

There are different technologies for providing positioning and navigation aid to AGVs. This project
focuses mainly on processing streaming images from cheap 360-degree cameras mounted on the AGVs.
The integrated 360-degree camera with a computer system to process streaming images shall provide at
least 2Hz positioning and navigation aid to the AGVs through a well-established interface.

Objectives and constraints

• Research 360-degree cameras and select best according to the design criteria.
• Research navigation aid interfaces for AGVs and document the interface requirements.
• Design and implement a positioning and navigation aid algorithm that can utilize 360-degree
images from a spherical camera.
• Design and implement a software prototype that can obtain 360-degree camera images, process
images according to the developed algorithm, and provide navigation aid through
selected/designed interface. The prototype implementation can also be carried out in simulation
environments such as https://cyberbotics.com.
• Design and implement required calibration and test scenarios.
• The positioning accuracy shall be 20 cm in a room of 10mx10m size, with a bearing angle
accuracy of 2 degrees (All accuracy values are given in 1 sigma RMS).

Milestones and Preliminary Design (Fall 2021, Week 10):

Deliverables • Design alternatives for positioning and bearing estimation methods and required camera
specifications, with associated decision analysis report.
• Decision analysis report for cheap camera solutions that provide stitched 360-degree images.
• Higher level design of system
Critical Design (Fall 2021, Finals)
• Software Requirement Analysis and Design
• Test Scenarios and Success criteria
• Analysis and Design Report
Implementation and Verification (Spring 2022)
• Source Codes of Software
• Test Reports
• Project Final Report

19
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Development of a 3D Satellite Orbit Simulator Software

Supervisor Asst. Prof. Dr. Murat Durmaz

Expected team size 3 students
Problem Definition Satellite orbit simulation is an important tool for space mission analysis, design and planning. Design of
single and multi-satellite missions includes precise orbit analysis and simulations of what-if cases. In
this project a Java based orbit simulation and reporting software will be developed using NASA
WorldWind. The simulation software shall also provide additional reports for different orbit
configurations.

Objectives and constraints Design and implement a 3D orbit simulator software that provides the following capabilities.

• The software shall provide a Graphical User Interface that includes

1) a 3D globe viewer based on WorldWind,
2) a form to define orbit using classical Keplerian elements, Position and Velocity
3) A list of satellites in simulation,
4) Simulation time display and adjustment interface.
• The software shall support the change of camera view into the nadir direction of a selected
satellite.
• The software shall support creation of sky view for a selected location on 3D globe.
• The software shall support calculation and display of visible region for the imaging sensor
deployed on a selected satellite. The software shall provide an interface for the definition of the
imaging sensor details.
• The software shall provide simultaneous simulation of orbits for at least 30 satellites without a
loss in responsiveness of the software.

Milestones and Preliminary Design (Fall 2021-2022, Week 10):

Deliverables • Alternatives for orbit calculation and simulation libraries with associated decision analysis
report.
• Higher level design of Software
Critical Design (Fall 2021-2022, Finals)
• Software Requirement Analysis and Design
• Test Scenarios and Success criteria
• Analysis and Design Report
Implementation and Verification (Spring 2021-2022).
• Source Codes of Software
• Test Reports
• Project Final Report

20
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Development of a QGIS plugin for time-series analyses and forecasting of the COVID-19 cases

Supervisor Dr. Engin Tunalı

Expected team size 4 students
Problem Definition Millions of people have been infected and have lost their lives due to the worldwide ongoing Coronavirus
(COVID-19) pandemic. It is of utmost importance to identify the future infected cases and the virus spread
rate for advance preparation in the healthcare services to avoid deaths. Accurately forecasting the spread
of COVID-19 is an analytical and challenging real-world problem to the Geo-Spatial community.

This project aims to analyze the time series data for several countries affected by the COVID-19 and model
spreading of the corona virus, and predict the impact to optimize the planning to manage the various
services and resources for the public by the governments. Series and Predictive analysis will be conducted
by employing Auto Regressive Integrated Moving Average (ARIMA) or Facebook’s (FB) Prophet
forecasting model.

Objectives and • Do research on ARIMA and FB-Prophet forecasting models and their applications.
constraints • Research available sources to extract the daily level up-to-date and continuous COVID-19 data.
• Design and implement required
o feature engineering steps,
o data processing algorithms and
o Graphical User Interface (GUI) of the QGIS plugin.
• Design and implement a software that extracts constantly updated daily cumulative COVID-19
data of different regions and predicts the future data and its accuracy by adopting the
aforementioned forecasting models.
• The plugin shall use an interactive global map (e.g., GIS dashboard) that being updated on a daily
basis and allow end users to pick different places and configure to see the prediction results and
general statistical info.

Milestones and Preliminary Design:

Deliverables • Testing ARIMA and Prophet models for time series forecasting of small COVID datasets.
• Appropriate model and parameters identification.
• Design the first prototype of GUI and determine possible outputs associated with decision
analysis report.
• Higher level design of models and plugin Interface.
Critical Design
• Plugin Requirement Analysis and Design
• Analysis and Design Report
Implementation and Verification
• Source Codes of Software, Test Reports, Final Report.

21
 Geomatics Engineering
Graduation Project Topics
Hacettepe University GMT403-GMT404 2021-2022

PROJECT DESCRIPTION
Project Title Development of a WEB Based Geodetic Toolbox that solve a variety of problems in Geodesy

Supervisor Dr. Engin Tunalı

Expected team size 4 students
Problem Definition Operations such as time and coordinate conversions and data cleaning are routine tasks in Geodesy.
Nevertheless, simple and efficient high-level functions to help those kinds of jobs are barely available, and
are developed, again and again, by each student, engineer for each new project. Despite the fact that countless
toolboxes already exist for scientific purposes, none really exists for geodetic - oriented purposes. The
objective of this project is to provide a simple but useful and efficient set of functions on WEB platform to
help Geomatics community to spend less time on the pre-processing steps and therefore gain in efficiency to
focus faster on their research.

Objectives and Design and implement a WEB Based Geodetic Toolbox that provides the following capabilities.
constraints • Time and coordinates transformation modules.
• Read, write and import geodetic files.
• File format conversion between different file standards .
• Download GNSS data/products from various servers.
The Toolbox shall support
• Single, file, batch coordinate conversion of multiple files.
• Recognizing several file formats of GIS.
• Parse GNSS Rinex, Broadcast and Precise Ephemeris data and store data to text file.

Milestones and Preliminary Design:

Deliverables • Determination of the platform and languages required for the project.
• Determining the modules to be applied and their abilities.
• Design the first prototype of GUI, modules and outputs associated with decision analysis report.
• Higher level design of tools and WEB page.
Critical Design
• Requirement Analysis and Design
• Analysis and Design Report
Implementation and Verification
• Source codes of the designed WEB page and defined geodetic modules,
• Test Reports,
• Final Report.

22