Bir Küçük Nesne Tespit Zorluğu Olarak Hava Görüntülerinden Araç Tespiti

Yıl 2021, Cilt: 4 Sayı: 1, 73 - 83, 15.01.2021

Ömer Er , Hasan Şakir Bilge

Öz

Hava görüntüleri üzerinde araç tespiti; istihbarat, keşif ve gözetleme açısından önemlidir. Ancak bu görev; düşük uzamsal çözünürlük, karmaşık arka plan, nesne üzerine düşen ışık/gölge farklılıkları ve nesnelerin çevre tarafından kamufle olması gibi sebeplerle zordur. Son zamanlarda geliştirilen CNN tabanlı ağlar umut vericidir ancak bu ağlar doğrudan küçük nesnelerin tespiti için yeterli değildirler ve ince ayara ihtiyaç duyarlar. Bu çalışmada daha hızlı RCNN algoritması ve görece büyük nesnelerin tespitinde başarısı kanıtlanmış ResNet ağı ile VEDAI veri kümesi üzerinde çalışılmıştır. Nesnelerin toplam görüntüdeki piksellerin 0.5×10−3’ü kadar az yer kapladığı görüntüler üzerinde başarım artırımı için daha hızlı RCNN algoritmasında değişiklikler ile çeşitli deneyler yapılmıştır. Deneyler sonucunda %74.9 ortalama hassasiyet elde etmenin mümkün olduğu gösterilmiştir.

Anahtar Kelimeler

Nesne Tespiti, Derin Öğrenme, Hava Görüntüleri

Destekleyen Kurum

ASELSAN A.Ş.

Kaynakça

• Liu C, Ding Y, Zhu M, Xiu J, Li M, Li Q. ”Vehicle Detection in Aerial Images Using a Fast Oriented Region Search and the Vector of Locally Aggregated Descriptors”. Sensors, 19(15), 3294, 2019.
• Dalal N., Triggs B. Histograms of oriented gradients for human detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition; San Diego, CA, USA. 20–25 June 2005.
• Lowe D. “Distinctive image features from scale-invariant keypoints”. International Journal of Computer Vision, 60, 91–110, 2004.
• Chang C, Lin C. LIBSVM: A library for support vector machines. ACM Trans. Intell. Syst. Technology, 389–396, 2011.
• Viola P, Jones M. Rapid object detection using a boosted cascade of simple features. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Kauai, USA, 2001.
• Felzenszwalb P, Girshick R, McAllester D, Ramanan D. Object detection with discriminatively trained part based models. IEEE Trans. Pattern Anal. Mach. Intell, 32, 1627–1645, 2010.
• Deng J, Berg A, Satheesh S, Su H, Khosla A, Li F. ImageNet Large Scale Visual Recognition Competition 2012, accessed on 10 July 2017.
• Krizhevsky A, Sutskever I, Hinton G. ImageNet classification with deep convolutional neural networks. Proceedings of the 25th International Conference on Neural Information Processing System, Lake Tahoe, NV, USA, 3–6 December 2012.
• Girshick R, Donahue J, Darrell T, Malik J. Rich feature hierarchies for accurate object detection and semantic segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA, 23–28 June 2014.
• Girshick R. Fast R-CNN. 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, 1440-1448, 2015
• Ren S, He K, Girshick R, Sun J. Faster R-CNN: Towards real-time object detection with region proposal networks. Neural Information Processing Systems (NIPS), 2015.
• Tayara H, Soo K, Chong K. Vehicle Detection and Counting in High-Resolution Aerial Images Using Convolutional Regression Neural Network. IEEE Access, 1-1, 2017
• Redmon J, Farhadi A. "An incremental improvement". CoRR, vol. 3, 2018.
• Redmon J, Divvala S, Girshick R, Farhadi A. "You only look once: Unified real-time object detection". IEEE Conference on ComputerVision and Pattern Recognition, 2016.
• Redmon J, Farhadi A. "YOLO9000: Better, Faster, Stronger". 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, 6517-6525, 2017
• Liu W, Anguelov D, Erhan D, Szegedy C, Reed S. "Ssd: Single shot multibox detector". European Conference on ComputerVision (ECCV), 2016.
• Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, Huang Z, Karpathy A, Khosla A, Bernstein M, Berg A. "Large Scale Visual Recognition Competition". International Computer Vision Journal, 2015.
• He K, Zhang X, Ren S, Sun J. "Deep residual learning for image recognition". IEEE Conference on Computer Vision and Pattern Recognition, 2016.
• Deng J, Dong W, Socher R. "Imagenet: A large-scale hierarchical image database". CVPR, 248-255, 2009.
• Lin T, Maire M, Belongie S, Hays J, Perona P, Ramanan D, Dollár P. "Microsoft COCO: common objects in context". ECCV,740-755, 2014.
• Razakarivony S, Jurie F. Vehicle detection in aerial imagery: A small target detection benchmark. J. Vis. Commun. Image Represent, 34, 187–203, 2016
• İnternet: Utah agrc URL:http://gis.utah.gov/. 2012.
• Zhong J, Lei T, Yao G. Robust Vehicle Detection in Aerial Images Based on Cascaded Convolutional Neural Networks. Sensors, 17(12), 2720, 2017.
• Sakla W, Konjevod G, Mundhenk T. "Deep Multi-modal Vehicle Detection in Aerial ISR Imagery". 2017 IEEE Winter Conference on Applications of Computer Vision, Santa Rosa, CA, 916-923, 2017.
• Uijlings J, Van de Sande K, Gevers T, Smeulders A. Selective search for object recognition. Int. J. Comput. Vision, 104,154–171, 2013
• Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv, 1409, 1556, 2014.
• Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, Erhan D, Vanhoucke V, Rabinovich A. “Going deeper with convolutions”. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1–9, 2015.
• Szegedy C, Ioffe S, Vanhoucke V. Inception-v4 inception-ResNet and the impact of residual connections on learning. CoRR, vol. abs/1602.07261, 2016
• Huster T, Deep learning for pedestrian detection in aerial imagery. In MSS Passive Sensors, 2016.
• Carlet J, Abayowa B. Fast vehicle detection in aerial imagery. CoRR, vol. Abs/1709.08666, 2017.
• Glorot X, Bengio Y. “Understanding the difficulty of training deep feedforward neural networks”. Journal of Machine Learning Research - Proceedings Track, 9, 249-256, 2010.