Yazar "Akcay, Ozgun" seçeneğine göre listele

Listeleniyor 1 - 5 / 5
  • [ X ]
    Öğe
    Building a semantic based public transportation geoportal compliant with the INSPIRE transport network data theme
    (Springer Heidelberg, 2014) Gunay, Arif; Akcay, Ozgun; Altan, Mehmet Orhan
    Interoperability is becoming essential for the today's geographic information systems. Geographic information is usually available as datasets stored in databases and accessible via GIS. However, these information sources are increasingly heterogeneous and show differences in data formats, database schema and object concepts. To satisfy the increased demand for the use and sharing of geographic data in common applications, considering the heterogeneity and the methods to support interoperability are required by the GIS community. The implementation of Spatial Data Infrastructure (SDI) such as the European INSPIRE (Infrastructure for Spatial Information in the European Community) enables accessibility and the sharing of geographic data and interoperability among the systems. In addition to this, traditional GIS systems perform spatial queries using a keyword-based method. However, this approach remains incapable of fully expressing the users' needs due to a lack of geographic concepts (semantics) in the dataset. Different terms may refer to similar concepts, while the same terms may refer to different concepts. This causes semantic heterogeneity in the dataset. In this context, the most promising approach is the implementation of geospatial semantics by means of ontology in the geospatial dataset to overcome this kind of ambiguity. The aim of this research is to investigate the implementation of INSPIRE as a SDI standard and the use of Semantic Technology to empower the traditional GIS approach. In this regard, a public transportation geoportal has been developed for the experimental investigation which uses a revised sample transportation data complaint with the INSPIRE Transport Network Data Theme and a developed Transportation Ontology Domain, including concepts, relationships and individuals to provide a semantic spatial search.
  • [ X ]
    Öğe
    Fiber quantity analysis of cementitious composite using multifocus imagery
    (Wiley, 2018) Akcay, Ozgun; Pekmezci, Bekir Yilmaz; Seker, Dursun Zafer; Tuncel, Erman Yigit; Atahan, Hakan Nuri
    After static and impact loading tests applied to PVA fiber reinforced cementitious composites, depending on the loading type and mixture properties of the composite, different fiber failure types are observed at the fractured surface of test samples. It is very hard to determine the pulled out fiber amount with conventional methods. The purpose of this study is to display these differences in a quantitative way using digital imaging processing techniques and exhibit the relation between absorbed energy and pulled out fiber area. Digital image processing might provide useful techniques to determine structure of the cracked surface of samples. Since close-up images of the failure surfaces of test samples are needed to examine thin and short fibers, from the point view of science of measurement and modeling from stereo images, digital photogrammetry does not produce prospective results appropriately. In this work, in order to reveal fibers, two complementary image-processing techniques were used successively. First, images, which obtained from macro photographic lenses at different focal lengths, were combined with multifocus imaging. Second, processed multifocused image is automatically clustered in order to extract fibers with maximum likelihood classification algorithm. Consequently, state of side-views of the fractured surfaces of composites is drawn precisely. The amount of pulled out fibers were estimated by using the area calculated. The relation between the absorbed energy and pulled fiber area was obtained. With increase of absorbed fracture energy, pulled out fiber amount for both static and dynamic tests increases. POLYM. COMPOS., 39:2126-2134, 2018. (c) 2016 Society of Plastics Engineers
  • [ X ]
    Öğe
    Landslide Fissure Inference Assessment by ANFIS and Logistic Regression Using UAS-Based Photogrammetry
    (Mdpi, 2015) Akcay, Ozgun
    Unmanned Aerial Systems (UAS) are now capable of gathering high-resolution data, therefore, landslides can be explored in detail at larger scales. In this research, 132 aerial photographs were captured, and 85,456 features were detected and matched automatically using UAS photogrammetry. The root mean square (RMS) values of the image coordinates of the Ground Control Points (GPCs) varied from 0.521 to 2.293 pixels, whereas maximum RMS values of automatically matched features was calculated as 2.921 pixels. Using the 3D point cloud, which was acquired by aerial photogrammetry, the raster datasets of the aspect, slope, and maximally stable extremal regions (MSER) detecting visual uniformity, were defined as three variables, in order to reason fissure structures on the landslide surface. In this research, an Adaptive Neuro Fuzzy Inference System (ANFIS) and a Logistic Regression (LR) were implemented using training datasets to infer fissure data appropriately. The accuracy of the predictive models was evaluated by drawing receiver operating characteristic (ROC) curves and by calculating the area under the ROC curve (AUC). The experiments exposed that high-resolution imagery is an indispensable data source to model and validate landslide fissures appropriately.
  • [ X ]
    Öğe
    Semantic Segmentation of High-Resolution Airborne Images with Dual-Stream DeepLabV3+
    (Mdpi, 2022) Akcay, Ozgun; Kinaci, Ahmet Cumhur; Avsar, Emin Ozgur; Aydar, Umut
    In geospatial applications such as urban planning and land use management, automatic detection and classification of earth objects are essential and primary subjects. When the significant semantic segmentation algorithms are considered, DeepLabV3+ stands out as a state-of-the-art CNN. Although the DeepLabV3+ model is capable of extracting multi-scale contextual information, there is still a need for multi-stream architectural approaches and different training approaches of the model that can leverage multi-modal geographic datasets. In this study, a new end-to-end dual-stream architecture that considers geospatial imagery was developed based on the DeepLabV3+ architecture. As a result, the spectral datasets other than RGB provided increments in semantic segmentation accuracies when they were used as additional channels to height information. Furthermore, both the given data augmentation and Tversky loss function which is sensitive to imbalanced data accomplished better overall accuracies. Also, it has been shown that the new dual-stream architecture using Potsdam and Vaihingen datasets produced 88.87% and 87.39% overall semantic segmentation accuracies, respectively. Eventually, it was seen that enhancement of the traditional significant semantic segmentation networks has a great potential to provide higher model performances, whereas the contribution of geospatial data as the second stream to RGB to segmentation was explicitly shown.
  • [ X ]
    Öğe
    The relationship between mucilage covered areas and chlorophyll-a concentration: The sea of Marmara case
    (Yildiz Technical Univ, 2022) Polat, Ahmet Batuhan; Sanli, Fusun Balik; Akcay, Ozgun
    Today, mucilage, or as it is known by the people, sea saliva, which directly affects the natural life and trade opportunities of the Marmara Sea, has more and more serious consequences day by day. The rapid detection of the spreading areas of the mucilage is of great importance for the cleaning works to be carried out. Remote sensing satellites have the great advantage of detecting the mucilage layer on the sea surface and studying its propagation. It has an important place in monitoring the change in that region over time and associating it with other ecological analysis, especially thanks to the ability to take images from the same region with frequent repetitions. Within the scope of this study, the mucilage distribution areas on the sea surface in the Dardanelles Strait, Gemlik Bay and Izmit Bay were investigated temporally and the concentration of chlorophyll-a, which is the main component organic matter of phytoplankton, and it is also the main cause of mucilage in the sea, was investigated and associated with mucilage areas. The satellite images used in the study were obtained from the Sentinel-2 multispectral optical satellite and the areas were determined by interpreting the images before and after mucilage propagation. Normalized Water Index (NDWI), a band index, was used to determine the mucilage area. As a result of the analysis, the mucilage rates in the sea were determined from the images on the dates used in the study. These rates were found as approximately 2.03% in the Dardanelles Strait, approximately 6.80% and 3.52% in the Gulf of Gemlik, where satellite images were examined at 10-day intervals, and approximately 7.44% in the Gulf of Izmit. When we look at the increase in the chlorophyll-a concentration in the sea for the same regions and dates, it is detected as about 0.84% in the Dardanelles Strait, 272.64% and 84.06% in the Gulf of Gemlik, and 8.22% in the Gulf of Izmit. When we look at the results, it is understood that the chlorophyll-a concentration and the spread of the mucilage layer are related to each other. In the light of these results, it is expected that using remote sensing techniques can be easily used for detecting mucilage, which is formed on the sea surface and spreads over large areas, and be a preliminary analysis study for rapid interventions.