- Remote-Sensing Image Classification
- Advanced Image Fusion Techniques
- Remote Sensing and Land Use
- Advanced Image and Video Retrieval Techniques
- Medical Image Segmentation Techniques
- Urban Green Space and Health
- Turkish Urban and Social Issues
- Advanced Vision and Imaging
- Urban Heat Island Mitigation
- Crime Patterns and Interventions
- Advanced Chemical Sensor Technologies
- Flood Risk Assessment and Management
- Image and Signal Denoising Methods
- Soil erosion and sediment transport
- Remote Sensing and LiDAR Applications
- Disaster Management and Resilience
- Building Energy and Comfort Optimization
- Public Administration and Governance
- Cultural Heritage Management and Preservation
- Energy and Environment Impacts
- Gender Studies and Social Issues
- Urban, Neighborhood, and Segregation Studies
- Soil Geostatistics and Mapping
- Crime, Illicit Activities, and Governance
- Sparse and Compressive Sensing Techniques
Izmir Institute of Technology
2021-2024
Kocaeli Üniversitesi
2012-2018
Hyperion Technologies (Canada)
2012
Middle East Technical University
2011
Terrain as a continuum can be categorized into landform units that exhibit common physical and morphological characteristics of land surface which may serve boundary condition for wide range application domains. However, heterogeneous views, definitions, applications on landforms yield incompatible nomenclature lacks interoperability. Yet, there is still room developing methods classification provide different disciplines with basis landscape description also commonsense to human insight....
Hyperspectral imaging provides high spectral resolution and thereby improved classification, detection, recognition capabilities with respect to standard systems. However, hyperspectral images generally have low spatial resolution, varying from a few tens of meters, resulting technical limitations such as platform data storing capacity satellite-to-ground transmission bandwidth. Spectral unmixing information on pixels in terms abundances pure signatures, without providing distribution at...
Abstract Vulnerability assessment and mapping play a crucial role in disaster risk reduction planning for adaptation to future earthquake. Turkey is one of the most at-risk countries earthquake disasters worldwide. Therefore, it imperative develop effective vulnerability at practically relevant scales. In this study, holistic index that addresses multidimensional nature was constructed. With aim representing as continuum across space, buildings were set smallest unit analysis. The study area...
Spectral unmixing is the process of identifying pure spectral signatures, called endmembers, from a hyperspectral data, and then expressing each pixel vector in terms fractional abundances these endmembers. Most endmember extraction methods literature use only information, whereas spatial composition data disregarded. Spatial preprocessing methods, that are motivated by assumption endmembers more likely to be located homogeneous regions instead transition areas, can alleviate this drawback...
Endmember extraction is the process of selecting pure spectral signatures materials from hyperspectral data. Most endmember methods in literature use only information, and disregard spatial composition image. Spatial-spectral preprocessing methods, motivated by assumption that endmembers are more likely to be located homogenous regions, can increase performance directing regions. However, such an approach generally results a failure extracting anomalous or scarce endmembers, which important...
Land management and planning is crucial for present future use of land the sustainability resources. Physical, biological cultural characteristics can be used to define Management Units (LMUs) that aid in decision making managing communicating information between different research application domains. This study aims describe classification ecologically relevant units are suitable management, conservation purposes. Relying on idea strong correlation landform potential landcover, a...
Kentsel Yeşil Alanların varlığı, kentlerde refah ve yaşam kalitesinin en önemli unsurlarından biri olarak kabul edilir. Alanlar, kent sakinleri için fiziksel aktivite, psikolojik rahatlama sağlar genel halk sağlığına katkıda bulunur. Alanlar oluşturmak, yoğun kentleşmeden dolayı etrafındaki flora faunadan izole olmuş kentlerin ekolojik çevreyi tekrar içine dahil etmesi çabasını da içerir, bu şekilde , işleyişi bütünlüğü desteklenmektedir.Kentsel Alanlarındiğer yararları ise, hava kirliliği,...
Abstract This paper presents an extensive evaluation of a novel intensity‐based image co‐registration method called luminance transformed differences ( LTD ) which is compared with current alternatives. The proposed suggests the use transform operation LTO to reduce between images that may well affect registration accuracy. Registration implemented in block‐based fashion order achieve non‐rigid transformation at sub‐pixel accuracy level. Sample from different sensors, exhibiting various...
İzmit İlçesi’nde yaz (Temmuz 2018 - Eylül 2018) ve kış mevsimlerinde (Kasım Ocak 2019) yüksek hacimli örnekleyici kullanılarak partikül madde (PM2.5) derişimleri ölçülmüştür. Mekânsal dağılımdaki farklılığı belirlemek için 14 noktada numune toplanmıştır. PM2.5 numuneleri, 24 saat boyunca 1,13 m3/dk bir akış hızında kuvars filtre kağıdı örnekleme noktası mevsimleri ortalama konsantrasyonu sırasıyla 22,11 ± 4,30 µg/m3 45,44 18,71 olarak en düşük konsantrasyonlarına aylarında rastlanmıştır. Kış...
Abstract. Along with urbanization, sealing of vegetated land and evaporation surfaces by impermeable materials, lead to changes in urban climate. This phenomenon is observed as temperatures several degrees higher densely urbanized areas compared the rural at fringe particularly nights, so-called Urban Heat Island. Island (UHI) effect related form, pattern building materials so far it associated meteorological conditions, air pollution, excess heat from cooling. UHI has negative influences on...
Climate Change, the biggest challenge of our era, has impacts highly variable and uneven across globe. Confirming this fact, projections suggest that Mediterranean basin is expected to face significant from climate change, showcasing region as a critical concern. The climatic changes in are characterized by shift towards aridity, i.e., ‘aridification’ an increment annual average temperature with associated decrease precipitation. Study area including İzmir Karaburun Peninsula located western...
Türkiye, yıkıcı deprem üretme potansiyeli yüksek fay hatlarının yer aldığı bir bölgede konumlanmaktadır. Tarih boyunca, Anadolu yarımadasında meydana gelen depremler büyük can ve mal kayıplarına sebep olmuştur. Bu bağlamda, nüfusun sanayileşmenin çok yoğun olduğu Marmara Bölgesinde risk değerlendirme çalışmalarının yoğunlaşması gerekli önlemlerin alınması önem arz etmektedir. çalışmada, 6306 sayılı "Afet Riski Altındaki Alanların Dönüştürülmesi Hakkında Kanun" kapsamında tanımlanan binaların...
The Urban Heat Island (UHI) effect is of critical concern for cities’ adaptation to climate change. UHI shows substantial intra-urban variation at the city microscale, causing disparities in thermal comfort and energy consumption. Therefore, air temperature assessment should be prioritized effective heat mitigation adaptation. However, meteorological stations’ spatial distribution far from meeting scale that its driving parameters operate. This limitation hampers demonstrating intra-city...
In this study we propose an automated fine registration of EO-1 Hyperion and IKONOS imagery. An intensity based that is area-based pixelwise performed to register given images divergent spatial spectral resolution. Two similarity measures are commonplace in image registration; NCC NMI, operation particular restoration; CTO adopted as error measure a novelty registration. We convinced with the performance efficiency compared other two common methods intensity-based
Intensity based image registration methods are widely used in fine geometric of multisensor images.Accordingly, for images that compared through translation templates, position where similarity measure is maximized assumed to indicate best registration.Image quality crucial importance especially studies have high accuracy requirements; e.g.image fusion, change detection, multichannel segmentation, and Digital Terrain Model (DTM) generation.Accuracy conventionally evaluated by means error...
Suç noktalarının basılı haritalar üzerine işaretlenerek gösterildiği eski yöntemler, Coğrafi Bilgi Sistemlerinin (CBS) ortaya çıkışıyla birlikte bilgisayar destekli grafik temsil yöntemlerine dönüşmüştür. Ayrıca, CBS durum ve olguların katman yapısıyla detaylı olarak edilmesini bu olgu durumların birbiri ile ilişkisini analiz etmeyi sağlar. suç haritaları mekanla ilgili durumları etmeye, suçun mekandaki dağılımını modellemeye ikisinin ederek suça neden olan mekânsal bağlamı anlamaya yardımcı...
Direnclilik kavrami, kentsel gelisim ve kentlerin surdurulebilirligi ile ilgili calismalarda giderek daha fazla yer edinmeye baslamistir. Dogal veya insan eliyle gerceklesen afetlerin dunya uzerinde yikici etkileri vardir. Ancak, kendi icinde bagintili karmasik bir yapiya sahip olan kentler, tum afetlere karsi cok kirilganlik gosterir. Kentlere has mekanizmalardan oturu, ozellikle buyuk kentlerde, afetlerle iliskili riskler buyuktur tam olarak kestirilemez. Bu durum, afet oncesi zarar...
Hyperspectral images having very high spectral resolution are subject to low spatial due technical constraints. Fusion of hyperspectral with multispectral produces information-content, and images. In this study for enhancement we propose unmixing-based approaches fusion Accordingly; end members extracted out adjusted they used derive abundances up a fused image. Method is applied on two datasets components as input one being synthetic other real dataset. produced better accuracy over...
In this paper, the robustness of measures that are used for image registration tested common distortions seen in images. Luminance Transform Operation (LTO) is proposed to be applications. Performance LTO compared Normilized Cross Correlation (NCC) and Mutual Information (MI) measures. Image process carried out between different wavelenght bands hyperspectral Non-linear luminance changes, noise rotation effects added target images, then performances NCC, MI their on corrupted Results...
This study intends to explore how various aspects of safety such as criminal acts, public order, and sense community contribute ‘perceived safety’. A questionnaire was applied 1050 individuals that comprise the sample set living within urban sprawl İzmit city. People were asked rate their evaluations on a issues in neighborhood they live, crime disorder, relationships. Investigating relevance these versus perceived is important understand characteristics an urbanizing city subject...
Bu çalışmada, Değirmendere beldesinde, Coğrafi Bilgi Sistemleri (CBS) temelli, depreme yönelik afet yönetimi zarar azaltma amacıyla bir sistem geliştirilmiştir. Geliştirilen için öncelikle bölgede yapılmış olan çalışmalar, 17 Ağustos 1999 depreminde oluşan zararlar ve sebepleri incelenmiş depremde yapıları etkileyen yerbilimleri verilerine dayanarak risk oranları verilerinin alt kategorilerinin etki oranlarına göre belirlenmiştir. Bölgenin tehlike analizinin yapmakta kullanılan bu verilerin...
Hyperspectral images compared to standard multispectral can resolve materials on earth with higher accuracy. However, due low spectral resolution of hyperspectral images, there exists the problem `mixed pixels'. Fusion high spatial is known yield rich information content products. In this study we propose unmixing-based fusion improve classification Accordingly; fused yielded results detail that image. Unmixing-based proved accuracy obtained from conventionally images. End member selection...
In the cases that spatial resolution of hyperspectral data is not sufficient, pixel vectors are expressed in terms abundances pure signatures, named as endmembers, with spectral mixture analysis. Most endmember extraction methods use only information, whereas pre-processing can increase performance by directing process to spatially homogeneous regions. However, this approach results a failure detecting anomaly endmembers. paper, two-way which provides high both anomalies and homogenous...
Spatial clustering of socioeconomic variables and socio-spatial segregation is a result the differentiation in choices living environment population groups that differ from each other their status. Typical examples spatial are residential areas low-income groups, commonly composed unqualified immigrants, certain city, higher status communities occupying places isolated rest society. The present study, conducted on an industrialized city was subject to mass immigration. Aim study evaluate if...