Recep Can

Researcher

image

Highly motivated researcher with a background in photogrammetry, computer vision and machine learning.
Ability to work interdisciplinary, quick adaption to new research topics and turn ideas into practice with
strong programming and problem solving skills.Good at time management.
Looking for new challenges to gain experience as researcher in the field of photogrammetry, geospatial data science and computer vision.


Work Experience

Researcher

TUBITAK UZAY | 2022 - Present

Remote Sensing Group

Research Assistant

Hacettepe University | 2020 - 2022

Department of Geomatics Engineering

Research Associate

April 2021 - Present

Evaluation of Skysat RPC Quality under ESA Earthnet Data Assessment Pilot (EDAP) Project Framework

Research Scholar

TUBITAK UZAY | 2019 - 2022

Geometric calibration of spaceborne linear array CCDs, developing sensor model and algorithms

R&D Team Member

LAMA Mobile App | 2018 - 2022

I am responsible for data quality control and visualization of the data on the web map and developing new ideas for the future of GeoCitSci.

Project Team Member

Gaziantep Bizim Şehir | 2018 - 2019

I worked for solving web development, 3D city modelling and 3D city visualization related problems.

Long Term Intern

UVM Systems TR | 2018 - Summer

3D City Modelling, 3D GIS

Summer Intern

Birlik Mapping | 2017

Producing Topographic Photogrammetric Maps, Orthophoto Generation, LiDAR Processing

Awards

ISPRS Best Young Author Award

XXIV ISPRS Congress

Journal Publications

Analysis of landslide susceptibility prediction accuracy with an event-based inventory: The 6 February 2023 Turkiye Earthquakes

Gizem Karakas, Erdinc Orsan Unal, Sinem Cetinkaya, Nazli Tunar Ozcan, Veysel Emre Karakas, Recep Can, Candan Gokceoglu, Sultan Kocaman

Landslide susceptibility assessment is a complex challenge explored by various scientists, but not fully resolved. In this study, we produced the landslide susceptibility map of a large region covering 38,500 km2 area in South-East Turkiye severely affected by the 6 February 2023 Kahramanmaras Earthquakes (Mw 7.7 and Mw 7.6) using an inventory produced in previous years. We employed random forest regression with a total of nine geomorphological and environmental features and evaluated the results using the co-seismic inventory with 2611 landslides compiled here. Although high accuracy was obtained from pixel-based assessments of test data split from the learning set, the independent validation set of co-seismic landslides showed that attention needs to be paid unseen features such as rare lithological units. Given the significant damage caused by latent hazards with the Kahramanmaras earthquakes, producing reliable inventories and precise landslide susceptibility maps is crucial for risk reduction and minimizing damages.
View Full-Text

A Comprehensive Assessment of XGBoost Algorithm for Landslide Susceptibility Mapping in the Upper Basin of Ataturk Dam, Turkey

Recep Can, Sultan Kocaman, Candan Gökçeoğlu

The success rate in landslide susceptibility mapping efforts increased with the advancements in machine learning algorithms and the availability of geospatial data with high spatial and temporal resolutions. Existing data-driven susceptibility mapping models are not globally applicable due to the high variability of landslide conditioning parameters and the limitations in the availability of up-to-date and accurate data. Among numerous applications, landslide susceptibility maps are essential for site selection and health monitoring of engineering structures, such as dams, for increasing their lifetime and to prevent from disastrous events caused by the damages. In this study, landslide susceptibility mapping performance of XGBoost algorithm was evaluated in a landslide-prone area in the upper basin of Ataturk Dam, which is a prime investment located in the southeast of Turkey. The study area has a size of 2718.7 km2 with an elevation difference of ca. 2000 m and contains 27 lithological units. EU-DEM v1.1 from the Copernicus Programme was used to derive the geomorphological features. High classification accuracy with area under curve value of 0.96 could be obtained from the XGBoost algorithm. According to the results, the main factors controlling the landslides in the study area are the lithology, altitude and topographic wetness index.
View Full-Text

A Convolutional Neural Network Architecture for Auto-Detection of Landslide Photographs to Assess Citizen Science and Volunteered Geographic Information Data Quality

Recep Can, Sultan Kocaman, Candan Gökçeoğlu

Several scientific processes benefit from Citizen Science (CitSci) and VGI (Volunteered Geographical Information) with the help of mobile and geospatial technologies. Studies on landslides can also take advantage of these approaches to a great extent. However, the quality of the collected data by both approaches is often questionable, and automated procedures to check the quality are needed for this purpose. In the present study, a convolutional neural network (CNN) architecture is proposed to validate landslide photos collected by citizens or nonexperts and integrated into a mobile- and web-based GIS environment designed specifically for a landslide CitSci project. The VGG16 has been used as the base model since it allows finetuning, and high performance could be achieved by selecting the best hyper-parameters. Although the training dataset was small, the proposed CNN architecture was found to be effective as it could identify the landslide photos with 94% precision. The accuracy of the results is sufficient for purpose and could even be improved further using a larger amount of training data, which is expected to be obtained with the help of volunteers.
View Full-Text

Conference Papers

CO-SEISMIC LANDSLIDE BASED VALIDATION OF SUSCEPTIBILITY MAPPING AFTER KAHRAMANMARAS EARTHQUAKES (FEB 6, 2023) IN AMANOS MOUNTAINS

G. Karakas, E. O. Unal, N. Tunar Ozcan, S. Cetinkaya, R. Can, C. Gokceoglu, and S. Kocaman

The quality of landslide susceptibility maps is often assessed using a part of learning data that represents geographical and land use characteristics over a quasi-fixed time. However, when validated with multi-temporal landslide inventories, more realistic insights on the susceptibility maps can be obtained. In addition, extreme events may trigger landslides in regions which are not considered as landslide-prone. The February 6, 2023, Kahramanmaras Earthquakes (Mw 7.7 and Mw 7.6), also known as the disaster of the century, triggered numerous landslides. Amanos Mountains located in southern Türkiye were also within the earthquake-affected area and had a very small amount of inventory recorded in official databases. The aim of this study was to evaluate the performance of the random forest method for producing landslide susceptibility maps. The official inventory of General Directorate of Mineral Research and Exploration (MTA) was used for map production. The resulting susceptibility map was assessed using the co-seismic landslide inventory produced in the study. The model’s performance evaluated using a part of the learning data yielded high accuracy expressed with area under receiver operating characteristics curve (AUC), precision, and recall values and F1 score using (AUC = 97%, recall = 97%, precision = 96%, F1 = 98%). However, multi-temporal evaluation with co-seismic landslides showed that 80% of the landslide pixels with moderate, high, and very high susceptibility levels could be predicted with the model. The results suggest that special attention should be given to features underrepresented in the inventory, such as low altitudes and types of lithology.

View Full-Text

Rock Mass Discontinuity Determination with Transfer Learning

I. Yalcin, R. Can, S. Kocaman, C. Gokceoglu

Rock mass discontinuity and orientation are among the important rock mass features. They are conventionally determined with scan-line surveys by engineering geologists in field, which can be difficult or impossible depending on site accessibility. Photogrammetry and computer vision techniques can aid to automatically perform these measurements, although variations in size, shape and appearance of rock masses make the task challenging. Here we propose an automated approach for the detection of rock mass discontinuities using deep learning and photogrammetric image processing methods. Two deep convolutional neural network (DCNNs) were implemented for this purpose and applied to basalts in Kizilcahamam Guvem Geosite near Ankara, Türkiye. Red-green-blue (RGB) band images of the site were taken from an off-the-shelf camera with 1.7 mm resolution and a 3D digital surface model and orthophotos were produced by using photogrammetric software. The discontinuities were delineated manually on the orthophoto and converted to masks. The first DCNN model was based on the open-source crack dataset consisting of a total of 11,298 road and pavement images, which were used to train the Resnet-18 model (Model-1). The second model (Model-2) was based on fine-tuning of Model-1 using the study data from Kizilcahamam. After fine-tuning, Model-2 was able to achieve high performance with a Jaccard Score of 88% on the test data. The results show high potential of the methodology for transfer learning with fine-tuning of a small amount of data that can be applied to other sites and rock mass types as well.

View Full-Text

Context-Based Satellite Image Search Engine

Kaan Kalkan, Zeynep Gökçe Aker, Recep Can

This study aims to develop a search engine that selects images similar to the areas selected by the user through a graphical interface using DSSP (Data Sharing Service Platform) data shared within APSCO (Asia-Pacific Space Cooperation Organization). The data received from the DSSP platform will be previously integrated into the system and divided into small image tiles. The user will be expected to select any of these tiles on the screen based on these areas. The developed system will detect satellite images that contain tiles similar to this tile in all data and the positions of these tiles. At this stage, the system will detect the areas similar to the selected tile with artificial learning methods and provide the return of the most similar areas to the user. These areas can be exemplified as tiles with buildings, forests, wind turbines or solar power plants.
View Full-Text

Land Cover Map Production with Deep Learning From Very High Resolution Satellite Imagery

İlyas Yalçın, Gizem Karakaş, Recep Can, Sultan Kocaman, Sebastien Saunier, Clement Albinet

Various applications such as monitoring the changes on Earth’s surface caused by natural hazards, rapid urbanization, etc., disaster management, planning of urban areas and agricultural lands, monitoring of natural resources have increased the need for up-to-date land cover maps and these maps must be produced with high temporal and spatial resolution. The data collection method used in the production of these maps is determined by taking into account the parameters such as the size of the study area, time and cost. For this purpose, the use of optical and radar satellite images, remote sensing techniques and artificial intelligence algorithms appear with an increasing frequency in the literature. MAXAR Technologies, one of the very high resolution satellite data providers, has started to offer images in High Definition (HD) format obtained from 30 cm Ground Sampling Distance (GSD) using advanced image processing methods. In this study, it was aimed to produce land cover maps with a deep learning approach using HD images of a selected study area. The defined classes were water surface, vegetation, asphalt, building, shade and open areas. The results showed that HD images are useful in land cover map production studies in urban areas and high classification accuracy can be obtained from the deep learning methods.
View Full-Text

Deep Learning Based Building Footfrint Extraction From Very High Resolution True Orthophotos And nDSM

Mehmet Büyükdemircioğlu, Recep Can, Sultan Kocaman Gökçeoğlu, Martin Kada

A challenging aspect of developing deep learning-based models for extracting building footprints from very high resolution (< 0.1 m) aerial imagery is the amount of details contained within the images. The use of convolutional neural networks (CNNs) to tackle semantic image segmentation has been shown to outperform conventional computer vision and machine learning approaches in various applications. Here, we investigated the performances of two different CNN architectures, U-Net and LinkNet by implementing them on various backbones and by using a number of building footprint vectors in a part of Turkey for training. The dataset includes red-green-blue (RGB) true orthophotos and normalized digital surface model (nDSM) data. The performances of the implemented methods were assessed comparatively by using the RGB data only and the RGB + nDSM. The results show that by adding nDSM as the fourth band to the RGB, the accuracy values obtained from the RGB only results were improved by 3.27% and 5.90% expressed in F1-Score and Jaccard (IoU) values, respectively. The highest accuracy reflected by the F1-Score of the validation data was 97.31%, while the F1-Score of the test data that was excluded from the model training was 96.14%. A vectorization process using the GDAL and Douglas-Peucker simplification algorithm was also performed to obtain the building footprints as polygons.
View Full-Text

A CNN based Flood Mapping Approach Using Sentinel-1 Data

Beste Tavus, Recep Can, Sultan Kocaman

The adverse effects of flood events have been increasing in the world due to the increasing occurrence frequency and their severity due to urbanization and the population growth. All weather sensors, such as satellite synthetic aperture radars (SAR) enable the extent detection and magnitude analysis of such events under cloudy atmospheric conditions. Sentinel-1 satellite from European Space Agency (ESA) facilitate such studies thanks to the free distribution, the regular data acquisition scheme and the availability of open source software. However, various difficulties in the visual interpretation and processing exist due to the size and the nature of the SAR data. The supervised machine learning algorithms have increasingly been used for automatic flood extent mapping. However, the use of Convolutional Neural Networks (CNNs) for this purpose is relatively new and requires further investigations. In this study, the U-Net architecture for multi-class segmentation of flooded areas and flooded vegetation was employed by using Sentinel-1 SAR data and altitude information as input. The training data was produced by an automatic thresholding approach using OTSU method in Sardoba, Uzbekistan and Sagaing, Myanmar. The results were validated in Ordu, Turkey and in Ca River, Vietnam by visual comparison with previously produced flood maps. The results show that CNNs have great potential in classifying flooded areas and flooded vegetation even when trained in areas with different geographical setting. The F1 scores obtained in the study for flood and flooded vegetation classes were 0.91 and 0.85, respectively.
View Full-Text

A CNN Architecture for Discontinuity Determination of Rock Masses with Close Range Images

Ilyas Yalcin, Recep Can, Sultan Kocaman,Candan Gökçeoğlu

Determination of discontinuities in rock mass requires scan-line surveys performed in in-situ that can reach up to dangerous and challenging dimensions. With the development of novel technological equipments and algorithms, the studies related to rock mass discontinuity determination remain up-to-date. Depending on the development of the Structure from Motion (SfM) method in the field of close-range photogrammetry, low-cost cameras can be used to produce 3D models of rock masses. However, the determination of rock mass discontinuity parameters must still be carried out manually on these models. Within the scope of this study, a Convolutional Neural Network (CNN) architecture is proposed to identify the discontinuities automatically as the first step for fully automated processing. The Kızılcahamam/Güvem Basalt Columns Geosite near Ankara, Turkey was determined as the study area. The orthophoto of this study area was produced using close-range photogrammetric methods and the training data was produced by manual mensuration. The dataset consists of labeled binary masks and images containing corresponding Red-Green-Blue (RGB) bands. Furthermore, the amount of data was increased by applying augmentation methods to the dataset. The U-Net architecture was used to detect rock discontinuities based on the produced orthophoto. The preliminary results presented here reveal that the discontinuity determination capability of the proposed method is high based on the visual assessments, while problems exist with image quality and discontinuity identification. In addition, considering the small size of the training dataset, the accuracy of the model would increase when a larger dataset could be employed.
View Full-Text

A WebGIS Framework for Semi-Automated Geodatabase Updating Assisted by Deep Learning

Recep Can, Sultan Kocaman, Ali Özgün Ok

The automation of geoinformation (GI) collection and interpretation has been a fundamental goal for many researchers. The developments in various sensors, platforms, and algorithms have been contributing to the achievement of this goal. In addition, the contributions of citizen science (CitSci) and volunteered geographical information (VGI) concepts have become evident and extensive for the geodata collection and interpretation in the era where information has the utmost importance to solve societal and environmental problems. The web- and mobile-based Geographical Information Systems (GIS) have facilitated the broad and frequent use of GI by people from any background, thanks to the accessibility and the simplicity of the platforms. On the other hand, the increased use of GI also yielded a great increment in the demand for GI in different application areas. Thus, new algorithms and platforms allowing human intervention are immensely required for semi-automatic GI extraction to increase the accuracy. By integrating the novel artificial intelligence (AI) methods including deep learning (DL) algorithms on WebGIS interfaces, this task can be achieved. Thus, volunteers with limited knowledge on GIS software can be supported to perform accurate processing and to make guided decisions. In this study, a web-based geospatial AI (GeoAI) platform was developed for map updating by using the image processing results obtained from a DL algorithm to assist volunteers. The platform includes vector drawing and editing capabilities and employs a spatial database management system to store the final maps. The system is flexible and can utilise various DL methods in the image segmentation.
View Full-Text

Deep Learning Based Roof Type Classification Using Very High Resolution Aerial Imagery

Mehmet Büyükdemircioğlu, Recep Can, Sultan Kocaman

Automatic detection, segmentation and reconstruction of buildings in urban areas from Earth Observation (EO) data are still challenging for many researchers. Roof is one of the most important element in a building model. The three-dimensional geographical information system (3D GIS) applications generally require the roof type and roof geometry for performing various analyses on the models, such as energy efficiency. The conventional segmentation and classification methods are often based on features like corners, edges and line segments. In parallel to the developments in computer hardware and artificial intelligence (AI) methods including deep learning (DL), image features can be extracted automatically. As a DL technique, convolutional neural networks (CNNs) can also be used for image classification tasks, but require large amount of high quality training data for obtaining accurate results. The main aim of this study was to generate a roof type dataset from very high-resolution (10 cm) orthophotos of Cesme, Turkey, and to classify the roof types using a shallow CNN architecture. The training dataset consists 10,000 roof images and their labels. Six roof type classes such as flat, hip, half-hip, gable, pyramid and complex roofs were used for the classification in the study area. The prediction performance of the shallow CNN model used here was compared with the results obtained from the fine-tuning of three well-known pre-trained networks, i.e. VGG-16, EfficientNetB4, ResNet-50. The results show that although our CNN has slightly lower performance expressed with the overall accuracy, it is still acceptable for many applications using sparse data.
View Full-Text

Landslide Susceptibility Mapping with Random Forest Model for Ordu, Turkey

Gizem Karakaş, Recep Can, Hakan A. Nefeslioğlu, Sultan Kocaman, Candan Gökçeoğlu

Landslides are among commonly observed natural hazards all over the world and can be quite destructive for infrastructure and in settlement areas. Their occurrences are often related with extreme meteorological events and seismic activities. Preparation of landslide susceptibility maps is important for disaster mitigation efforts and to increase the resilience. The factors effective on landslide susceptibility map production depend mainly on the topography, land use and the geological characteristics of the region. The up-to-date and accurate data needed for extracting the effective parameters can be obtained by using photogrammetric techniques with high spatial resolution. Data driven ensemble methods are being increasingly used for landslide susceptibility map production and accurate results can be obtained. In this study, regional landslide susceptibility map of a landslide-prone area in a part of Ordu Province in northern Turkey is produced using topographic and lithological parameters by employing the random forest method. An actual landslide inventory delineated manually by geologists using the produced orthophotos and the digital terrain model (DTM) is used for training the model. The results show that an accuracy of 83% and precision of 92% can obtained from the data and the random forest method. The approach can be applied for generation of regional susceptibility maps semi-automatically.
View Full-Text

Development of a CitSci and Artificial Intelligence Supported GIS Platform for Landslide Data Collection

Recep Can, Sultan Kocaman, Candan Gökçeoğlu

Geospatial data are fundamental to understand the relationship between the geographical events and the Earth dynamics. Although the geospatial technologies aid geodata collection, the increasing possibilities yield new application areas and cause even a greater demand. Considering the increment in data quantity and diversity, to be able to work with the data, they must be collected, stored, analysed and presented with the help of specifically designed platforms. Geographical Information Systems (GIS) with mobile and web support are the most suitable platforms for these purposes. On the other hand, the location-enabled mobile, web and geospatial technologies empowered the rise of the citizen science (CitSci) projects. With the CitSci, mobile GIS platforms enable the data to be collected from almost any location. As the size of the collected data increases, considering automatic control of the data quality has become a necessity. Integrating artificial intelligence (AI) with the CitSci based GIS designs allows automatic quality control of the data and helps eliminating data validation problem in CitSci. For this reason, the purpose of the present study is to develop a CitSci and AI supported GIS platform for landslide data collection because landslide hazard mitigation efforts require landslide susceptibility, hazard and risk assessments. Especially, landslide hazard assessments are necessary the time of occurrence of a landslide. Although this information is crucial, it is almost impossible to collect time of occurrence in regional hazard assessment efforts. Consequently, use of CitSci for this purpose may provide valuable information for landslide hazard assessments.
View Full-Text

Projects

LAMA GeoCitSci

WebGIS Developer

Creating web map that shows photographs obtained from LAMA mobile app with their correct position. Working for to automate controlling data quality using deep learning and creating web GIS application.

Gaziantep Bizim Şehir

Technical Support

Transforming models and data that comes from architectural team to georeferenced 3D city models, visualization of that models in CityGRID Scout, technical support for software related problems.

Bizim Şehir Gaziantep Web Sayfası