METU CV & RS LAB

METU COMPUTER VISION AND REMOTE SENSING RESEARCH GROUP

The Computer Vision and Remote Sensing Research Group targets to conduct research on the cutting-edge topics of computer vision and remote sensing. Research projects vary from human perspective oriented evaluation of systems to development of new algorithm and methods for technical issues.  Additionally, experiments on remote and nondestructive image acquisition methods are performed in the laboratory in order to observe the difficulties faced in real data problems. Our research is directed mainly towards the following areas:

  • Image Processing
  • Hyperspectral – Multispectral Imaging
  • Omnidirectional Vision
  • Pattern Recognition
  • Machine Vision
  • Video Surveillance
  • GPU Programming
  • Virtual Reality
  • Computer Graphics
  • Human-Computer Interaction

Equipment

  • Headwall A Series VNIR Linescan camera
  • JAI Multispectral Areascan camera
  • Thorslab Quartz Tungsten Halogen Lamp
  • Horizontal and vertical camera slider
  • Parabolic and hyperbolic mirror systems for omnidirectional imaging 
  • Imagingsource monochrome and color cameras

Group Members

Principal Investigator: Assoc. Prof. Dr. Alptekin Temizel, Prof. Yasemin Yardımcı Çetin

Students

  • Fatih Ömrüuzun (Ph.D. Candidate)
  • Umut Çınar (Ph.D. Candidate)
  • Okan Bilge Özdemir (Ph.D. Candidate)
  • Didem Özışık Başkurt (Ph.D. Candidate);

Alumni

  • Dr. Yalın Baştanlar - Assist Prof. at Izmir Institute of Technology
  • Dr. Musa Ataş - Assist.Prof. at Siirt University
  • Dr. Ersin Karaman - Assist.Prof. at Atatürk University
  • Çiğdem Beyan - PhD candidate in the School of Informatics at the University of Edinburgh
  • Püren Güler - Ph.D. student at KTH, Stockholm, Sweden
  • Deniz Emeksiz - Software Specialist at Innova

Publications

  • Y. Baştanlar, Temizel A., Yardımcı Y., & Sturm P. Multi-view Structure-from-Motion for Hybrid Camera Scenarios. Image and Vision Computing, 30(8), 557–572, 2012.
  • M. Atas, Yardımcı Y., & Temizel A. "A New Approach to Aflatoxin Detection in Chili Pepper by Machine Vision. Computers and Electronics in Agriculture," 87(C), 129-141, 2012.
  • Y. Baştanlar, Temizel A., Yardımcı Y., & Sturm P. "Multi-view Structure-from-Motion for Hybrid Camera Scenarios. Image and Vision Computing," 30(8), 557–572, 2012.
  • Baştanlar, Y., Temizel A., & Yardımcı Y. (2010). Improved SIFT Matching for Image Pairs with a Scale Difference. IET Electronics Letters. 46(5)
  • H.D.Sevim, Y. Yardımcı Çetin, D. O. Baskurt, A novel method to detect shadows on multispectral images, SPIE Remore Sensing, Edinburgh, UK, September(2016.)
  • D. O. Baskurt, Y. Gür, F.Ömrüuzun, Y.Yardımcı Çetin, Gas detection by using transmittance estimation and segmentation approaches, SPIE Remore Sensing, Edinburgh, UK, September (2016.)
  • F. Ömrüuzun, B.Demir, L. Bruzzone and Y. Yardimci Cetin, content based hyperspectral ımage retrıeval usıng bag of endmembers ımage descrıptors, ıeee whıspers, los angeles, august (2016.)
  • Okan Bilge Özdemir, Yasemin Yardımcı Çetin, Hilal Soydan , Yasemin Yardımcı Çetin, H. Şebnem Düzgün;  Hyperspectral Unmixing Based Vegetation Detection with Segmentation, IEEE IGARSS, Beıjıng, China, July 2016.
  • Teke, Mustafa, and Yasemin Yardimci. "Classification of crops using multitemporal hyperion images." Agro-Geoinformatics (Agro-geoinformatics), 2015 Fourth International Conference on. IEEE, 2015.
  • Omruuzun, F., Baskurt, D. O., Daglayan, H., & Cetin, Y. Y. (2015, October). Utilizing hyperspectral remote sensing imagery for afforestation planning of partially covered areas. In SPIE Remote Sensing (pp. 96432N-96432N). International Society for Optics and Photonics.
  • F Omruuzun, D Baskurt, YY Cetin. DTW based signal alignment for enhancing CO2 detection in MWIR hyperspectral imagery, IEEE Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing 2015
  • Omruuzun, F., Baskurt, D. O., Daglayan, H., & Cetin, Y. Y. (2015, June). Shadow removal from VNIR hyperspectral remote sensing imagery with endmember signature analysis. In SPIE Sensing Technology+ Applications(pp. 94821F-94821F). International Society for Optics and Photonics.
  • Omruuzun, F , Cetin, Y. Y. “Endmember signature based detection of flammable gases in LWIR hyperspectral images”, Proc. SPIE 9486, Advanced Environmental, Chemical, and Biological Sensing Technologies XII, 948612 (May 13, 2015); doi:10.1117/12.2182060
  • Omruuzun, Fatih, and Yasemin Yardimci Cetin. "Gas detection in longwave infrared hyperspectral imagery and black body effect compensation." Signal Processing and Communications Applications Conference (SIU), 2015 23th. IEEE, 2015.
  • Ozdemir, Okan Bilge, et al. "Signature based vegetation detection on hyperspectral images." Signal Processing and Communications Applications Conference (SIU), 2015 23th. IEEE, 2015.
  • Baskurt, Didem Ozisik, Fatih Omruuzun, and Yasemin Yardimci Cetin. "Hyperspectral unmixing based analysis of forested areas." Signal Processing and Communications Applications Conference (SIU), 2015 23th. IEEE, 2015.
  • F Omruuzun, YY Cetin. Content based hyperspectral image retrieval: A systematic review, Hyperspectral Imaging and Applications Conference, 2014
  • Fatih Omruuzun, Okan Bilge Ozdemir, Yasemin Yardimci Çetin. METU SPEL: Development of a New Spectral Signature Library For Food Products, Advances in Mass Data Analysis of Images and Signals in Medicine, Biotechnology, Chemistry and Food Industry, 9th International Conference, MDA 2014, 2014
  • Çaglar Senaras, Ekin Gedik, Yasemin Yardimci. A novel dynamic thresholding and categorizing approach to extract water objects from VHR satellite images. IGARSS 2014: 4934-4937
  • Çaglar Senaras, Ekin Gedik, Yasemin Yardimci Çetin. A new dynamic thresholding method for detection of water regions in multispectral VHR images. SIU 2014: 1512-1515
  • Okan Bilge Ozdemir, Yasemin Yardimci Çetin. Improvement of hyperspectral classification accuracy with limited training data using meanshift segmentation. SIU 2014: 1794-1797
  • Ekin Gedik, Umut Çinar, Ersin Karaman, Yasemin Yardimci, Ugur Halici, S. Kubilay Pakin, Hamza Ergezer. Automatic water canal detection in multispectral satellite images. SIU 2013: 1-4
  • Okan Bilge Ozdemir, Yasemin Yardimci Çetin. The effect of training data on hyperspectral classification algorithms. SIU 2013: 1-4 

Projects

  • Hyperspectral Image Analysis of Archeological Remaining, METU BAP, 2015-Ongoing
  • İGT-TUYGUN, Undersecretariat for Defense Industries of Turkey, 2014
  • Hyperspectral Image Improving and Object Classification (HİGİN), Ministry of Industry and Commerce of Turkey,2013
  • Material Detection on Hyperspectral Images, METU BAP, 2013
  • Anomaly Detection for Crowded Environment Video Surveillance Applications, Tubitak 1001, 2014, Link: ii.metu.edu.tr/node/484
  • Abandoned Object Detection With Information Fusion of Thermal and Visible Band Images
  • Video Surveillance on GPU using CUDA
  • CamDroid: A Web Services Based Live Camera Display Engine for Android Platform
  • 3D Reconstruction and Web Based Virtual Tour for Cultural Heritage of Aegean Region
  • Camera Sabotage Discovery for Video Survelliance Applications
  • Food Safety with Non-Invasive Techniques
  • Virtual Crowd Generation
  • Out-the-window Scene Properties in PC-based Helicopter Simulators
  • Evaluation of Visual Cues of Three Dimensional Virtual Environments for Helicopter Simulators

External Collaborators

  • OGAM
  • ITI-Certh
  • Argedor Inc
  • Headwall Photonics