Research Article
BibTex RIS Cite

İnsansız Hava Araçları ve Uydu Görüntülerinden Elde Edilen Veri Seti ile Havaalanlarının Tespitinin Yapılmasında SSD ve Faster R-CNN Algoritmalarının Karşılaştırılması

Year 2020, Issue: 19, 643 - 658, 31.08.2020
https://doi.org/10.31590/ejosat.742789

Abstract

Günümüzde görüntü işleme çalışmalarında birçok faklı sektörde, özellikle sağlık, üretim ve askeri alanlarda, doğrudan insan yaşantısında çeşitli amaçlarla kullanılmaya başlanmıştır. Derin öğrenme algoritmalarının gelişmesi ve bilgisayarlı görüde kullanılmaya başlanması özellikle askeri alandaki kritik hedef, önemli konum ve stratejik bölge tespiti gibi çalışmalara hız kazandırmıştır. Bu çalışmada Airport olarak adlandırılan havaalanlarının, uçak iniş pistleri üzerinden tespiti gerçekleştirilmiştir. Hem orta ve yüksek irtifalı insanız hava araçlarından hem de uydu görüntüleri kullanılarak eğitim, test ve değerlendirme veri setleri oluşturulmuştur. Tespit yapılması sürecinde SSD-Single Shot Multibox algoritması ve Faster R-CNN algoritması yeniden eğitilerek kullanılmıştır. Her iki algoritmanın sonuçları doğruluk oranı, duyarlılık, özgüllük, yanlış pozitif oranı, yanlış negatif oranı, doğru tahmin oranı, F puanı, hata oranı, sonuç ve eğitim süresi gibi değerlendirme kriterleri kapsamında değerlendirilmiştir. Değerlendirme veri seti üzerinde; SSD mimarisi ile %76,61 doğruluk oranıyla, Faster R-CNN mimarisinde ise %99,52 doğruluk oranı ile görüntü tespit sonucu elde edilmiştir. Söz konusu çalışma ile iki mimariden hangisinin insansız hava araçları ve uydu görüntülerinde kritik bölge tespitinde ne derece başarılı olduğu ortaya çıkarılmıştır.

References

  • Girshick, R., Donahue, J., Darrell, Malik, T. J., Berkeley, U. C. «Rich feature hierarchies for accurate object detection and semantic segmentation. » 2014.
  • Girshick, R., «Fast R-CNN». Proceedings of the IEEE international conference on computer vision, 2015, ss 1440-1448.
  • Kamran, F., Shahzad, M., Shafait, F. «Automated Military Vehicle Detection From Low-Altitude Aerial Images. » 2018 Digit. Image Comput. Tech. Appl. 2018. ss 1–8.
  • Xiaozhu, X., «Object Detection of Armored Vehicles Based on Deep Learning in Battlefield Environment.» 2017. ss 1569–1571.
  • Hsu, C., Chang, C., Lin, C. «A Practical Guide to Support Vector Classification.» vol. 1, no. 1. 2016. ss 1–16.
  • Github, «Tensorflow detection model zoo.» , https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/detection_model_zoo.md, [Reach date: 18.11.2019].
  • Polat, H., Mehr, H. D., & Cetin, A. «Diagnosis of chronic kidney disease based on support vector machine by feature selection methods. » Journal of medical systems, 41(4), 2017, ss. 55.
  • Lasko, T. A., Bhagwat, J. G., Zou, K. H., & Ohno-Machado, L., «The use of receiver operating characteristic curves in biomedical informatics. Journal of biomedical informatics. » 38(5), 2005, ss.404-415.
  • Metz, C. E., «Receiver operating characteristic analysis: a tool for the quantitative evaluation of observer performance and imaging systems. » Journal of the American College of Radiology, 3(6), 2006, ss 413-422.
  • Dirican, A., «Tanı testi performanslarının değerlendirilmesi ve kıyaslanması.» Cerrahpaşa Tıp Dergisi, 32(1), 2001, ss. 25-30.
  • Fawcett, T., An introduction to ROC analysis. Pattern Recogn. Lett.27:861–874, 2006. doi:10.1016/j.patrec.2005.10.010.
  • Lane, J. E., & Gantley, M. J., «Utilizing Complex Systems Statistics for Historical and Archaeological Data». Journal of Cognitive Historiography, 3(1-2), 2017, ss 68-92.
  • Chen, K., «How to interpret “loss” and “accuracy” for a machine learning model.», Web Site: https://stackoverflow.com/questions/34518656/how-tointerpretloss-and-accuracy-for-a-machine-learning-model, 2017, [Reach Date: 20.12.2019].
  • Google, https://datasetsearch.research.google.com/ [Reach Date: 20.12.2019].

Comparison of SSD and Faster R-CNN Algorithms to Detect the Airports with Data Set Which Obtained From Unmanned Aerial Vehicles and Satellite Images

Year 2020, Issue: 19, 643 - 658, 31.08.2020
https://doi.org/10.31590/ejosat.742789

Abstract

Today, image processing has been used in many different sectors, especially in health, production and military fields, for various purposes directly in human life. The development of deep learning algorithms and starting to use of computer vision has accelerated the studies such as critical target, important location and strategic region determination especially in the military field. In this study, the airport has been determined on the landing runways. Training, test and evaluation data sets were created by using both medium and high-altitude unmanned air vehicles and satellite images. SSD-Single Shot Multibox algorithm and Faster R-CNN algorithm were used by re-training during the determination process. The results of both algorithms were evaluated within the extend of evaluation criteria such as accuracy, sensitivity, specificity, false positive rate, false negative rate, positive pred value, F score, error rate, result and training time. The image detection accuracy with SSD algorithm was 76,61%, with Faster R-CNN algorithm the image detection accuracy was 99.52% according to valuation dataset. With this study, which of the two architectures has been revealed to be successful in determining critical areas in unmanned aerial vehicles and satellite images.

References

  • Girshick, R., Donahue, J., Darrell, Malik, T. J., Berkeley, U. C. «Rich feature hierarchies for accurate object detection and semantic segmentation. » 2014.
  • Girshick, R., «Fast R-CNN». Proceedings of the IEEE international conference on computer vision, 2015, ss 1440-1448.
  • Kamran, F., Shahzad, M., Shafait, F. «Automated Military Vehicle Detection From Low-Altitude Aerial Images. » 2018 Digit. Image Comput. Tech. Appl. 2018. ss 1–8.
  • Xiaozhu, X., «Object Detection of Armored Vehicles Based on Deep Learning in Battlefield Environment.» 2017. ss 1569–1571.
  • Hsu, C., Chang, C., Lin, C. «A Practical Guide to Support Vector Classification.» vol. 1, no. 1. 2016. ss 1–16.
  • Github, «Tensorflow detection model zoo.» , https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/detection_model_zoo.md, [Reach date: 18.11.2019].
  • Polat, H., Mehr, H. D., & Cetin, A. «Diagnosis of chronic kidney disease based on support vector machine by feature selection methods. » Journal of medical systems, 41(4), 2017, ss. 55.
  • Lasko, T. A., Bhagwat, J. G., Zou, K. H., & Ohno-Machado, L., «The use of receiver operating characteristic curves in biomedical informatics. Journal of biomedical informatics. » 38(5), 2005, ss.404-415.
  • Metz, C. E., «Receiver operating characteristic analysis: a tool for the quantitative evaluation of observer performance and imaging systems. » Journal of the American College of Radiology, 3(6), 2006, ss 413-422.
  • Dirican, A., «Tanı testi performanslarının değerlendirilmesi ve kıyaslanması.» Cerrahpaşa Tıp Dergisi, 32(1), 2001, ss. 25-30.
  • Fawcett, T., An introduction to ROC analysis. Pattern Recogn. Lett.27:861–874, 2006. doi:10.1016/j.patrec.2005.10.010.
  • Lane, J. E., & Gantley, M. J., «Utilizing Complex Systems Statistics for Historical and Archaeological Data». Journal of Cognitive Historiography, 3(1-2), 2017, ss 68-92.
  • Chen, K., «How to interpret “loss” and “accuracy” for a machine learning model.», Web Site: https://stackoverflow.com/questions/34518656/how-tointerpretloss-and-accuracy-for-a-machine-learning-model, 2017, [Reach Date: 20.12.2019].
  • Google, https://datasetsearch.research.google.com/ [Reach Date: 20.12.2019].
There are 14 citations in total.

Details

Primary Language English
Subjects Engineering
Journal Section Articles
Authors

Muhammed Taha Zeren 0000-0001-5615-0751

Sabahattin Kerem Aytulun 0000-0002-4688-0408

Yasin Kırelli 0000-0002-3605-8621

Publication Date August 31, 2020
Published in Issue Year 2020 Issue: 19

Cite

APA Zeren, M. T., Aytulun, S. K., & Kırelli, Y. (2020). Comparison of SSD and Faster R-CNN Algorithms to Detect the Airports with Data Set Which Obtained From Unmanned Aerial Vehicles and Satellite Images. Avrupa Bilim Ve Teknoloji Dergisi(19), 643-658. https://doi.org/10.31590/ejosat.742789