DSpace Arşivi :: by Yazar "Karas, I.R." değerine göre listeleniyor

Yazar "Karas, I.R." seçeneğine göre listele

Listeleniyor 1 - 20 / 43

A 3D campus information system-initial studies
(International Society for Photogrammetry and Remote Sensing, 2013) Kahraman, I.; Karas, I.R.; Alizadehasharfi, B.; Abdul-Rahman, A.
This paper discusses the method of developing Campus Information System. The system can handle 3D spatial data within desktop and web environment. The method consists of texturing of building facades for 3D building models and modeling 3D Campus Information System. In this paper, some of these steps are carried out; modelling 3D buildings, toggling these models on the terrain and ortho-photo, integration with a geo-database, transferring to the CityServer3D environment by using CityGML format and designing the service, etc. In addition to this, a simple but novel method of texturing of building façades for 3D city modeling that is based on Dynamic Pulse Function(DPF) is used for synthetic and procedural texturing. DPF is very fast compared to other photo realistic texturing methods. Last but not least, it is aimed to present this project on web using web mapping services. This makes 3D analysis easy for decision makers.
3D indoor navigation prototype for smartphones
(Conference Chairs of 3DGeoInfo 2014 in Karlsruhe, 2014) Ortakci, Y.; Karas, I.R.; Rahman, A.A.
Nowadays, there are a lot of multi-storey, complex and huge buildings in the cit ies especially in metropolises. These buildings are almost like a small city with their tens of floors, hundreds of corridors and rooms and passages. Sometimes people lost their way in these huge buildings. Due to the size and complexity of these buildings, people need guidance to find their way to the destination in these buildings. In this study, a mobile application has been developed to visualize a pedestrian's indoor location as 3D in his/her smartphone. This mobile application has the characteristics of a prototype for indoor navigation systems. While the pedestrian is walking on his/her way, the smartphone will guide the pedestrian with the photos of indoor environment on the route, arrow marks and text information. As a future plan, an RFID (Radio-Frequency Identification) technology can be integrated to the system to detect the location of the pedestrian during his/her tour in the building. By this way, the system will navigate the users more accurately as a real-time navigation system. Copyright © (2014) by the corresponding authors of the papers.
A 3D-GIS implementation for realizing 3D network analysis and routing simulation for evacuation purpose
(Kluwer Academic Publishers, 2013) Atila, U.; Karas, I.R.; Rahman, A.A.
The need for 3D visualization and navigation within 3D-GIS environment is increasingly growing and spreading to various fields. When we consider current navigation systems, most of them are still in 2D environment that is insufficient to realize 3D objects and obtain satisfactory solutions for 3D environment. One of the most important research areas is evacuating the buildings with safety as more complex building infrastructures are increasing in today's world. The end user side of such evacuation system needs to run in mobile environment with an accurate indoor positioning while the system assist people to the destination with support of visual landscapes and voice commands. For realizing such navigation system we need to solve complex 3D network analysis. The objective of this paper is to investigate and implement 3D visualization and navigation techniques and solutions for indoor spaces within 3D-GIS. As an initial step and as for implementation a GUI provides 3D visualization of Corporation Complex in Putrajaya based on CityGML data, stores spatial data in a Geo-Database and then performs complex network analysis under some different kind of constraints. The GUI also provides a routing simulation on a calculated shortest path with voice commands and visualized instructions which are intended to be the infrastructure of a voice enabled mobile navigation system in our future work. © 2013 Springer-Verlag Berlin Heidelberg.
Application of exploratory spatial techniques in the identification of tourism hotspots in the aegean region of Turkey
(International Society for Photogrammetry and Remote Sensing, 2020) Rafique, A.; Karas, I.R.; Abujayyab, S.K.M.; Khan, A.A.; Demiral, E.
Exploratory Spatial Analysis Techniques (ESDA) have become popular to identify the spatial association of different variables in many fields of natural and social sciences. The application of Global Moran's I statistics enables us to provide visual insights of spatial data. It helps to detect spatial patterns and hotspots of an activity or process, based on spatial autocorrelation. This study aims to investigate the spatial dependence of domestic and inbound tourist arrivals to 123 cities of all eight provinces of the Aegean Region of Turkey. For analysis, city-level data about domestic and inbound tourist arrivals during 2015-2019 is collected from the Turkish Ministry of Culture and Tourism and is converted to logarithm form to avoid any skewness. The Arc GIS and GeoDa programs are employed for the analysis of spatial autocorrelation and visualization of hotspots of tourist flows in the regions. The results of the study reveal that tourist flows in the region are concentrated in the coastal areas, while inland cities receive an insufficient number of tourists. The hotspots of tourist flow are located mostly in the coastal towns of the provinces of Izmir, Aydin, and Mugla. The study is significant in the provision of useful information regarding resource allocation to the tourism hotspots and the implication of sustainable tourism policy to better utilization of tourism potential. © 2020 International Society for Photogrammetry and Remote Sensing. All rights reserved.
AUTOMATED PREDICTION SYSTEM for VEGETATION COVER BASED on MODIS-NDVI SATELLITE DATA and NEURAL NETWORKS
(International Society for Photogrammetry and Remote Sensing, 2019) Abujayyab, S.K.M.; Karas, I.R.
Around the world, vegetation cover functioning as shelter to wildlife, clean water, food security as well as treat large part of air pollution problem. Accurate predictive data early warn and provide knowledge for decision makers to reduce the effects of changes in vegetation cover. In this paper, an automated prediction system was developed to forecast vegetation cover. Prediction system based on moderate satellite data spatial resolution and global coverage data. The tools of system automate processing Moderate Resolution Imaging Spectroradiometer (MODIS) images and training neural networks (NN) model based on 60,000 observations to forecast future density of Normalized Difference Vegetation Index (NDVI). Zonguldak data, located in north of Turkey as dense vegetation cover area utilized as case study for system application. This system significantly facilitates predictive process for users than previous long and complex models. © 2020 Authors.
Automatic generation of 3D networks in cityGML and design of an intelligent individual evacuation model for building fires within the scope of 3D GIS
(Kluwer Academic Publishers, 2014) Atila, U.; Karas, I.R.; Turan, M.K.; Rahman, A.A.
Designing 3D navigation systems requires addressing solution methods for complex topologies, 3D modelling, visualization, topological network analysis and so on. 3D navigation within 3D-GIS environment is increasingly growing and spreading to various fields. One of those fields is evacuation through the shortest path with safety in case of disasters such as fire, massive terrorist attacks happening in complex and tall buildings of today’s world. Especially fire with no doubt is one of the most dangerous disaster threatening these buildings including thousands of occupants inside. This chapter presents entire solution methods for designing an intelligent individual evacuation model starting from data generation process. The model is based on Multilayer Perceptron (MLP) which is one of the most preferred artificial neural network architecture in classification and prediction problems. We focus on integration of this model with a 3D-GIS based simulation for demonstrating an individual evacuation process. © Springer International Publishing Switzerland 2014.
A COMPARISON of TREE-BASED ALGORITHMS for COMPLEX WETLAND CLASSIFICATION USING the GOOGLE EARTH ENGINE
(International Society for Photogrammetry and Remote Sensing, 2021) Jamali, A.; Mahdianpari, M.; Karas, I.R.
Wetlands are endangered ecosystems that are required to be systematically monitored. Wetlands have significant contributions to the well-being of human-being, fauna, and fungi. They provide vital services, including water storage, carbon sequestration, food security, and protecting the shorelines from floods. Remote sensing is preferred over the other conventional earth observation methods such as field surveying. It provides the necessary tools for the systematic and standardized method of large-scale wetland mapping. On the other hand, new cloud computing technologies for the storage and processing of large-scale remote sensing big data such as the Google Earth Engine (GEE) have emerged. As such, for the complex wetland classification in the pilot site of the Avalon, Newfoundland, Canada, we compare the results of three tree-based classifiers of the Decision Tree (DT), Random Forest (RF), and Extreme Gradient Boosting (XGB) available in the GEE code editor using Sentinel-2 images. Based on the results, the XGB classifier with an overall accuracy of 82.58% outperformed the RF (82.52%) and DT (77.62%) classifiers. © Author(s) 2021. CC BY 4.0 License.
CORRECTION of FAULTY LINES in MUSCLE MODEL, to BE USED in 3D BUILDING NETWORK CONSTRUCTION
(Copernicus GmbH, 2012) Karas, I.R.; Atila, U.; Abdul-Rahman, A.
This paper describes the usage of MUSCLE (Multidirectional Scanning for Line Extraction) Model for automatic generation of 3D networks in CityGML format (from raster floor plans). MUSCLE (Multidirectional Scanning for Line Extraction) Model is a conversion method which was developed to vectorize the straight lines through the raster images including floor plans, maps for GIS, architectural drawings, and machine plans. The model allows user to define specific criteria which are crucial for acquiring the vectorization process. Unlike traditional vectorization process, this model generates straight lines based on a line thinning algorithm, without performing line following-chain coding and vector reduction stages. In this method the nearly vertical lines were obtained by scanning the images horizontally, while the nearly horizontal lines were obtained by scanning the images vertically. In a case where two or more consecutive lines are nearly horizontal or nearly vertical, raster data become unmanageable and the process generates wrongly vectorized lines. In this situation, to obtain the precise lines, the image with the wrongly vectorized lines is diagonally scanned. By using MUSCLE model, the network models are topologically structured in CityGML format. After the generation process, it is possible to perform 3D network analysis based on these models. Then, by using the software that was designed based on the generated models, a geodatabase of the models could be established. This paper presents the correction application in MUSCLE and explains 3D network construction in detail.
Design of a route guidance system with shortest driving time based on genetic algorithm
(2011) Atila, U.; Karas, I.R.; Gologlu, C.; Yaman, B.; Orak, I.M.
Nowadays, with the advancement of the technology on mobile devices, route guidance systems that assist drivers on the traffic have become widespread in daily life. For an accurate routing, a route guidance system should consider the effectual factors of traffic flow such as density and allowable velocity limits of the roads. With the increase of effectual factors and amount of nodes in road network, the computational cost increases. It is not proper to find exact optimal solution in real time for the road networks with excessive number of nodes using some well known deterministic methods such as Dijkstra's algorithm on navigation systems using mobile devices with limited processing speed and memory capacity. This paper proposes a route guidance system and a Genetic Algorithm (GA) approach applied on this routing system to find the shortest driving time. Excluding classical methods, a gene search method of chromosomes named "firstmatched-genes" on crossover operation had been introduced. The efficiency of the genetic algorithm was tested by applying on the networks with different sizes and a mobile application on the traffic network of Ankara was presented.
Design of an intelligent individual evacuation model for high rise building fires based on neural network within the scope of 3D GIS
(Copernicus GmbH, 2013) Atila, U.; Karas, I.R.; Turan, M.K.; Rahman, A.A.
One of the most dangerous disaster threatening the high rise and complex buildings of today's world including thousands of occupants inside is fire with no doubt. When we consider high population and the complexity of such buildings it is clear to see that performing a rapid and safe evacuation seems hard and human being does not have good memories in case of such disasters like world trade center 9/11. Therefore, it is very important to design knowledge based realtime interactive evacuation methods instead of classical strategies which lack of flexibility. This paper presents a 3D-GIS implementation which simulates the behaviour of an intelligent indoor pedestrian navigation model proposed for a self -evacuation of a person in case of fire. The model is based on Multilayer Perceptron(MLP) which is one of the most preferred artificial neural network architecture in classification and prediction problems. A sample fire scenario following through predefined instructions has been performed on 3D model of the Corporation Complex in Putrajaya (Malaysia) and the intelligent evacuation process has been realized within a proposed 3D-GIS based simulation. © 2013 Copernicus. All Rights Reserved.
Developing a 3D routing instruction engine for indoor environment
(Springer Verlag, 2017) Karas, I.R.; Atila, U.; Demiral, E.
The need for 3D visualization and navigation within 3D-GIS environment is increasingly growing and spreading to various fields. When we consider current navigation systems, most of them are still in 2D environment that is insufficient to realize 3D objects and obtain satisfactory solutions for 3D environment. For realizing such a 3D navigation system we need to solve complex 3D network analysis. The objective of this paper is to investigate and implement 3D visualization and navigation techniques and develop 3D routing instruction engine for indoor spaces within 3D-GIS. As an initial step and as for implementation a Graphical User Interface provides 3D visualization based on CityGML data, stores spatial data in a Geo-Database and then performs complex network analysis. By using developed engine, the GUI also provides a routing simulation on a calculated shortest path with voice commands and visualized instructions. © Springer Nature Singapore Pte Ltd. 2017.
DEVELOPMENT of IOT ENABLED GLOBAL TRACKING SYSTEM and MOBILE APPLICATION for PEOPLE with ALZHEIMER'S DISEASE
(International Society for Photogrammetry and Remote Sensing, 2021) I°leri, K.; Duru, A.; Karas, I.R.
Alzheimer's is a degenerative disease meaning that it gets worse with time. Memory loss, speaking problems, wandering, and getting lost are some of the signs of the disease. The risk of wandering results in high demand for extensive monitoring solutions for the patients suffering from the disease. Tracking solutions are crucial, especially for family members and caregivers, so researchers develop new wearable tracking devices to overcome missing patients. GPS technology can provide location data with high accuracy, but it is not sufficient to use only by itself. Thus, a more extensive solution should be provided. In this paper, a mobile wearable tracking device that can provide data to the mobile application through internet has been developed for patient tracking purposes. © Author(s) 2021. CC BY 4.0 License.
Dijkstra algorithm interactive training software development for network analysis applications in GIS
(2011) Karas, I.R.; Demir, S.
Process of route optimization is one of the basic applications of Network Analyses in Geographic Information Systems. In mathematical background of network analysis applications are graph theory and graph algorithms. Primary graph algorithm employed in process of route optimization is Dijkstra's Algorithm. Dijkstra's Algorithm is placed on the top of linear methods which yield exact solutions. Geographic Information Systems analyses such as the shortest route, the shortest duration and route with the least traffic are solved through Dijkstra's Algorithm. In this study, an interactive training software program, developed for educational use in Geographic Information Systems and Graph Theory classes at postgraduate degree, is introduced. This software provides students with the opportunity to use Dijkstra's Algorithm on graphs which they have designed by themselves and teaches details of algorithm, its working principles and structure of data to them, step by step, through interactive messages and graphics. © Sila Science.
Effective machine learning techniques for brain pathology classification on mr images
(American Institute of Physics, 2024) Mahmood, R.M.; Ramaha, N.T.A.; Karas, I.R.
Since a brain tumor is essentially a collection of aberrant tissues, it is crucial to classify tumors of the brain using MRI before beginning therapy. Tumor segmentation and classification using machine learning from brain MRI scans are well-known to be challenging and important endeavors. Machine learning has the potential to be used in diagnostics, preoperative planning, and postoperative evaluations. Furthermore, it is crucial to get accurate measurements of the tumor's location on an MRI of the brain. The development of machine learning models and other technologies will let radiologists detect malignancies without having to cut into patients. Pre-processing, skull stripping, and tumor segmentation are the steps in detecting a brain tumor and measurement (size and form). After a certain period, CNN models get overfitted because of the large number of training images used to train them. That is why this study uses deep CNN to transfer learning. CNN-based Relu architecture and SVM with fused retrieved features via HOG and LPB are used to classify brain MRI tumors (glioma or meningioma). The methods' efficacy is measured by precision, recall, F-measure, and accuracy. This study showed that the accuracy of the SVM with combined LBP with HOG is 97%, and the deep CNN is 98%. © 2024 Author(s).
The fifth international conference on smart city applications: Preface
(International Society for Photogrammetry and Remote Sensing, 2020) Karas, I.R.; Ahmed, M.B.; Abdelhakim, A.B.; Ane, B.K.
[No abstract available]
A genetic algorithm approach for finding the shortest driving time on mobile devices
(2011) Karas, I.R.; Atila, U.
Recently, with the increasing interest in using handheld devices, the application of navigation systems that provide driving information to the drivers has become widespread in daily life. An efficient route guidance system should consider the influential factors of traffic flow such as traffic density and allowable velocity limits of the roads. As the number of influential factors and amount of nodes in road network increase, the computational cost increases. On navigation systems, using handheld devices with limited processing speed and memory capacity, it is not feasible to find the exact optimal solution in real-time for the road networks with excessive number of nodes using deterministic methods such as Dijkstra algorithm. This paper proposes a Genetic Algorithm approach applied to a route guidance system to find the shortest driving time. Constant length chromosomes have been used for encoding the problem. It was found that the mutation operator proposed in this algorithm provided great contribution to achieve optimum solution by maintaining the genetic diversity. The efficiency of the genetic algorithm was tested by applying it on the networks with different sizes. ©2011 Academic Journals.
Genetic algorithm-aided routing on 3D dynamic networks
(International Society for Photogrammetry and Remote Sensing, 2010) Atila, U.; Karas, I.R.
3D network analysis for indoor provides strong decision support for users in searching optimal routes on applications such as emergency services, transportation, security and visitor guiding. Genetic algorithm is used to solve non-linear problems with complicated constraints. Therefore, the implementation of genetic algorithm into route finding algorithms is needed. This paper explains the demand using genetic algorithm approach on dynamic network routing problems especially for 3D navigation. Abilities of genetic algorithm is investigated as a search strategy and necessitates of genetic algorithm on use for 3D dynamic network routing is presented.
GEOSPATIAL MACHINE LEARNING DATASETS STRUCTURING and CLASSIFICATION TOOL: CASE STUDY for MAPPING LULC from RASAT SATELLITE IMAGES
(International Society for Photogrammetry and Remote Sensing, 2019) Abujayyab, S.K.M.; Karas, I.R.
Remote sensing satellite images plays a significant role in mapping land use/land cover LULC. Machine learning ML provide robust functions for satellite image classification. The objective of this paper is to extend the capability of GIS specialists in geospatial area with minimum knowledge in computer science to easily perform ML satellite image classification. A framework consisting 7 stages established. Tools of steps developed in two programing environments, which are ArcGIS for geospatial datasets structuring and Anaconda for ML training and classification. During the development, authors constrained to reduce the complexity of big data of satellite images and limited memory of computers to make tools available for implementation in PC. In addition, automation and improving the performance accuracy. TensorFlow-Keras library employed to perform the classification using neural networks. A case study using RASAT satellite image in Ankara-Turkey utilized to perform the analysis. The developed classifier gained 80% performance accuracy. The complete RASAT satellite image processed and smoothly classified based on blocks methods. The developed tools successfully tested and applied in geospatial area and can be effectively execute in PC by GIS specialist. © 2019 S. K. M. Abujayyab.
Handling massive data size issue in buildings footprints extraction from high-resolution satellite images
(Springer, 2020) Abujayyab, S.K.M.; Karas, I.R.
Building information modelling BIM is relying on plenty of geospatial information such as buildings footprints. Collecting and updating BIM information is a considerable challenge. Recently, buildings footprints automatically extracted from high-resolution satellite images utilizing machine learning algorithms. Constructing required training datasets for machine learning algorithms and testing data is computationally intensive. When the analysis performs in large geographic areas, researchers are struggling from out of memory problems. The requirement of developing improved, fit memory computation methods for accomplishing this computation is urgent. This paper targeting to handling massive data size issue in buildings footprints extraction from high-resolution satellite images. This article established a method to process the spatial raster data based on the chunks computing. Chunk-based decomposition decomposes raster array into several tiny cubes. Cubes supposed to be small enough to fit into available memory and prevent memory overflow. The algorithm of the method developed using Python programming language. Spatial data and developed tool were prepared and processed in ArcGIS software. Matlab software utilized for machine learning. Neural networks implemented for extracting the buildings’ footprints. To demonstrate the performance of our approach, high-resolution Orthoimage located in Tucson, Arizona state in American United States was utilized as a case study. Original image was taken by UltraCamEagle sensor and contained (11888 columns, 11866 rows, cell size 0.5 foot, 564,252,032 pixels in 4 bands). The case image contained (1409 columns, 1346 rows, and 7586056 pixels in 4 bands). The full image is impossible to be handled in the traditional central processing unit CPU. The image divided to 36 chunks using 1000 rows and 1000 columns. Full analysis spent 35 min using Intel Core i7 processor. The output performance accuracy of the neural network is 98.3% for testing dataset. Consequences demonstrate that the chunk computing can solve the memory overflow in personal computers during buildings footprints extraction process, especially in case of processing large files of high-resolution images. The developed method is suitable to be implemented in an affordable lightweight desktop environment. In addition, building footprints extracted effetely and memory overflow problem bypassed. Furthermore, the developed method proved the high quality extracted buildings footprints that can be integrated with BIM applications. © Springer Nature Switzerland AG 2020.
A Hybrid Method for Road Marking Detection
(Praise Worthy Prize, 2022) Mazouzi, A.; Yssaad, S.R.; Karas, I.R.
Very recently, the problems posed in the context of road safety have taken off with the growing number of accidents on the roads. These accidents result in property damage and death in several cases. Nowadays, it is no longer a question of settling for a simple conventional and ordinary road safety, but an automatic road safety is widely desired. This study fits precisely into this context and proposes, first and in terms of computer vision, a combination of two types of approaches, which are by appearance and structural. Indeed the appearance approach is presented by the statistical classifier, while the structural one is given by the set of the following methods: the detection of the segments using the algorithm of the semi local analysis, the combined Kalman filter with predictor to estimate the road line, and the layout adjustment by applying the principle of Minimum Mean Square Error (MMSE). This method is tested on several sample images and videos. It has given a good quality results with very short processing times. © 2022 Praise Worthy Prize S.r.l.-All rights reserved.

Yazar "Karas, I.R." seçeneğine göre listele

Sayfa Başına Sonuç

Sıralama seçenekleri