Firat Uray

Mr. Firat Uray

orcid

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 November 2021 in Survey Review
Verified
Journal Article
Closed access
AVAbdullah Varlik
Abdullah Varlik
FUFirat Uray
Firat Uray
The classification of LIDAR point clouds has always been a challenging task. Classification refers to label each point in different categories, such as ground, vegetation or building. The success of deep learning techniques in image processing tasks have encouraged researchers to use deep neural networks for classification of LIDAR point clouds. In this paper, we proposed a U-Net based architecture capable of classifying LIDAR data. The results indicated that our network model achieved an average F1 score of 91% over all three classes (ground, vegetation and building) for our best model.

1 April 2018 in Geomatik
Verified
Journal Article
Open access
PDF file
FUFirat Uray
Firat Uray
AVAbdullah VARLIK
Abdullah VARLIK
AMAzim Meti̇n
Azim Meti̇n
3B (Üç Boyutlu) kent modellerinin en önemli özelliği farklı mekânsal bilgilerin aynı ortamda bütünleştirilip gösterimine ve karmaşık kent modellerinin oluşturulup bunların yönetimine olanak sağlamasıdır. 3B sanal kent modelleri, arazi modelleri, bina modelleri, bitki modelleri, yollar gibi ulaşım sistemlerini içeren 3B ortamlar ve coğrafi tabanlı şehir verilerinin gösterimini içermektedir. 3B binalar için ölçek kavramı, LoD (Level of Detail) ayrıntı düzeyleriyle ifade edilmektedir. Her bir LoD belirli bir genelleştirme düzeyini gösterir. Bu çalışmanın ana amacı; kent tasarımı yapan tüm disiplinlerin gereksinim duyduğu farklı LoD seviyelerinde üç boyutlu kent modellerini, lazer tarayıcılardan elde edilen nokta bulutu verileri de kullanılarak üretimlerini araştırmaktır. Bu metodun nasıl ve ne şekilde uygulanacağı, yapılacak entegrasyon sonunda nasıl ve ne kalitede bir ürün elde edileceği gibi temellerin tartışılmasıdır. Bu sayede söz konusu verilerin kullanılabilirliği ve uygunluğu araştırılacaktır. Bu amaçla Konya’nın simgesi olma özelliğini taşıyan İnce Minareli Medrese seçilmiştir. Çalışma sonunda elde edilen model gerçek ölçeğinde fotogrametrik verilerin kullanıldığı yersel lazer tarayıcı destekli üç boyutlu yapı modelidir.

1 January 2015 in Procedia Earth and Planetary Science
Verified
Journal Article
Open access
FUFirat Uray
Firat Uray
AMA. Metin
A. Metin
AVA. Varlik
A. Varlik
Turkey is one of the countries hosting many historical monuments. Many different cultures left rich traces of civilization in Anatolia. Keeping historical artifacts and documentation of these works is very important for the protection of this heritage. The technological progress changed the methods of surveying applications used nowadays. The changes in the methods of surveying applications reflected in the building survey applications too. The traditional surveying methods were replaced by digital photogrammetry and then by laser scanning methods. This work is contains the building survey of the Ulu Beden tower which is a part of Diyarbakir walls. Diyarbakir is a province in southeastern Turkey and in the past, it was home for many civilizations. Hittites, Persia, Macedonian and Seleucid Greeks, Rome, Byzantium, Arabs, Seljuk Empire, Mongol Empire and Ottoman Empire are among these civilizations leaving the traces in Diyarbakir. Diyarbakir wall is the second longest wall after the Great Wall of China. Hundreds of years after the construction of the walls, it is necessary to protect and restore it due to the corruption and demolition over time. In total, 45 stations were built for an architectural survey with laser scanner and the gathered point cloud data were combined together within computer environment to obtain a 3D model of the structure.

1 January 2016 in Harita Teknolojileri Elektronik Dergisi
Verified
Journal Article
Closed access
AVAbdullah Varlik
Abdullah Varlik
FUFirat Uray
Firat Uray
AMAzim Meti̇n
Azim Meti̇n
Alanındaki gelişmeleri takip etmek, meslek kuruluşları ve bireylerin ulusal ve uluslararası gelişimlerine katkıda bulunmak, bu alanlarda Türkçe bir kaynak oluşturmak amacıyla yazılmış araştırma makaleleri, kısa makaleler, teknik notlar ve teknolojik tartışma yazılarını yayınlayan hakemli bir elektronik dergidir.

31 December 2024 in Necmettin Erbakan Üniversitesi Fen ve Mühendislik Bilimleri Dergisi
Verified
Journal Article
Open access
PDF file
FUFirat Uray
Firat Uray
AVAbdullah VARLIK
Abdullah VARLIK
In recent years, airborne and terrestrial laser scanning systems have become increasingly popular for obtaining geospatial information. High-quality 3D point clouds are used for a wide range of applications, with digital terrain models (DTMs) representing one of these products. Light Detection and Ranging (LiDAR) data is widely used in producing digital elevation models (DEM), which are fundamental elements of planning applications. Removing non-ground objects from LiDAR point clouds is the main problem of the DTM production workflow. Triangulated irregular network (TIN) densification is a classical technique among LiDAR filtering algorithms. In this study, a simple filtering algorithm entitled simple TIN densification (sTIN) is proposed, which is derived from classic TIN densification. The performance of sTIN is tested with three filters, namely, the adaptive triangulated irregular network, the simple morphological filter and the improved progressive TIN densification. The proposed algorithm has an average type I error rate of 5.6%, an average type II error rate of 10.42% and an average total error rate of 8.2%. In addition, the strengths and weaknesses of the algorithms are examined with regards to the root-mean-square error (RMSE) values and visual analyses of DTMs.

25 March 2026 in Turkish Journal of Remote Sensing
Verified
Journal Article
Closed access
FUFirat Uray
Firat Uray
AVAbdullah VARLIK
Abdullah VARLIK
LiDAR point cloud classification has always been a difficult task. Deep learning (DL) models have recently been widely used in computer vision studies, thanks to rapid advances in machine learning technology. In recent years, deep neural networks (DNN) have been used for classification and segmentation of LiDAR point clouds. We proposed a Fully Convolutional Network (FCN) architecture that can classify LiDAR data. This workflow is built around converting a three-dimensional (3D) LiDAR point cloud to a single two-dimensional (2D) multi-channel image. We tested the performance of our neural network model on the DALES airborne LiDAR dataset. The results showed that our network model performed well when compared to the LAStools software, with an average F1 score of 86% for our best model across all three classes.