Open Access Open Access  Restricted Access Subscription Access

Scene based Classification of Aerial Images using Convolution Neural Networks


Affiliations
1 Department of Computer Science & IT, University of Jammu, J&K, India
2 Department of Electronics, University of Jammu, J&K, India
 

The advent of computer vision and evolution of high-end computing in remote sensing images have embellish various researchers for unprecedented development in remotely sensed aerial images. The requirement to extract essential information stimulated anatomization of aerial images for its usefulness. Deep learning provides state of the art solutions for widely explored visual recognition system and has emerged as an evolutionary area, being applicable to large scale image processing applications. Convolutional Neural Networks (CNNs), an essential component of deep learning algorithms consists of increasing the depth and connections in the processing layers to learn various features of data at different abstract levels. In this paper, we present an outlook for classifying and extracting the features of aerial images using CNN. We propose a CNN architecture based on various parameters and layers for classification. CNN has been evaluated on two publicly available aerial data sets: UC Merced Land Use and RSSCN7. Experimental results show that the proposed CNN architecture is competent and efficient in terms of accuracy as performance evaluation parameter in comparison with conventional classifiers like Bag of Visual Words (BOVW).

Keywords

CNN, Deep Learning, Feature Extraction, Image Classification.
User
Notifications
Font Size

  • Mou L, Zhu X, IM2HEIGHT: Height estimation from single monocular imagery via fully residual convolutional-deconvolutional network, arXiv preprint (2018) arXiv:1802.10249.
  • Sun Y, Liang D, Wang X, Tang X, DeepID3: face recognition with very deep neural networks, arXiv preprint (2015) arXiv:1502.00873.
  • Samaniego L, Schulz K, Supervised classification of agricultural land cover using a modified k-NN technique and landsat remote sensing imagery, J Remote Sens, 1 (2009) 875–895.
  • Huang L, Liu B, Li B, Guo W, Yu W, Zhang Z, Yu W, A dataset dedicated to sentinel-1 ship interpretation, IEEE J Sel Top Appl Earth Obs Remote Sens, 11(1) (2018) 195–208.
  • Yang Y, Newsam S, Bag-of-visual-words and spatial extensions for land-use classification, Proc of SIGSPATIAL International Conference on Advances in Geographic Information Systems, 2010, 270–279.
  • Mahajan P, Abrol P, Lehana P K, Effect of blurring on identification of aerial images using convolution neural networks, Lect Notes Electr Eng, 597 (2020) 469–484.
  • Arel I, Rose D C, Karnowski T P, Deep machine learning - a new frontier in artificial intelligence research, IEEE Comput Intell Mag 5(4) (2010) 13–18.
  • Chen Y, Lin Z, Zhao X, Wang G, Gu Y, Deep learning-based classification of hyperspectral data, IEEE J Sel Top Appl Earth Obs Remote Sens 7(6) (2014) 2094–2107.
  • Ghamisi P, Plaza J, Chen Y, Li J, Plaza A, Advanced spectral classifiers for hyperspectral images, IEEE Geosci and Remote Sens Mag 5(1) (2017) 8–32.
  • Bishop C M, Pattern Recognition and Machine Learning, Springer, 2007, New York.
  • Bengio Y, Learning deep architectures for AI, Found Trends Mach Learn 2(1) (2009) 1–127.
  • Bouvrie J, Notes on convolutional neural networks, 2006.
  • Sevo I, Avramovic A, Convolutional neural network based automatic object detection on aerial images, IEEE Geosci and Remote Sens Lett 13(5) (2016) 740 – 744.
  • Kobayashi F K, Mattos A B, Gemignani B H, Macedo M G, Experimental analysis of citrus tree classification from UAV images, IEEE International Symposium on Multimedia 2019.
  • Ioffe S, Szegedy C, Batch normalization: accelerating deep network training by reducing internal covariate shift, arXiv preprint (2015) arXiv:1502.03167.
  • Zou Q, Ni L, Zhang T, Wang Q, Deep learning based feature selection for remote sensing scene classification, IEEE Geosci Remote Sens Lett, 12(11) (2015) 2321–2325.
  • Castelluccio M, Poggi G, Sansone C, Verdoliva L, Land use classification in remote sensing images by convolutional neural networks improving spatial, arXiv preprint (2015) arXiv:1508.00092v1.
  • Nogueira K, Miranda W O, Santos J A, Improving spatial feature representation from aerial scenes by using convolutional networks, Proc SIBGRAPI Conference on Graphics, Patterns and Images, 2015, 289–296.
  • Krizhevsky, Sutskever I, Hinton G E, ImageNet classification with deep convolutional neural networks, Adv Neur Inf Proces Syst, (2012) 1106–1114.
  • Simonyan K, Zisserman A, Very deep convolutional networks for large-scale image recognition. arXiv preprint (2015) arXiv:1409.1556.
  • Tayara H, Soo KG, Chong KT, Vehicle detection and counting in high-resolution aerial images using convolutional regression neural network, IEEE Access, 6 (2017) 2220–2210.
  • Schlemper J, Caballero J, Hajnal J V, Price A N, Rueckert D, A deep cascade of convolutional neural networks for dynamic mr image reconstruction, IEEE Trans Med Imaging, 37(2) (2018) 491–503.
  • Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, Erhan D, Vanhoucke V, Rabinovich A, Going deeper with convolutions, Proc IEEE Conference on Computer Vision and Pattern Recognition, 2015, 1–9.
  • Sambyal P, Abrol P, Lehana P, Optimization of light switching pattern on large scale using genetic algorithm, Int J Sci Tech Adv, 3(1), (2017) 19–23.
  • Xie L, Yuille A, Genetic CNN.arXiv preprint (2017) arXiv:1703.01513.
  • Kavukcuoglu K, Sermanet P, Boureau Y L, Gregor K, Mathieu M, LeCun Y, Learning convolutional feature hierarchies for visual recognition unsupervised deep feature, Proc ACM Conference on Advances in Neural Information Processing Systems, 2010, 1090–1098.
  • Romero A, Gatta C, Valls G C, Unsupervised deep feature extraction for remote sensing image classification, IEEE Trans Geosci Remote Sens, 54(3) (2016) 1349–1362.
  • Cheriyadat A M, Unsupervised feature learning for aerial scene classification, IEEE Trans on Geosci Remote Sens, 52(1) (2014) 429–451.
  • Cao F, Yang Z, Ren J, Ling WK, Extreme sparse multinomial logistic regression: a fast and robust framework for hyperspectral image classification, arXiv preprint (2017) arXiv:1709.02517.
  • Aydogdu M F, Celik V, Demirci M F, Comparison of three different cnn architectures for age classification, Proc IEEE International Conference on Semantic Computing, 2017, 372–377.
  • Nagi J, Ducatelle F, Di Caro G A, Ciresan D, Meier U, Giusti A, Nagi F, Schmidhuber J, Gambardella L M, Maxpooling convolutional neural networks for vision based hand gesture recognition, In Proc IEEE International Conference on Signal and Image Processing Applications, 2011, 343–349.
  • LeCun Y, Kavukcuoglu K, Farabet C, Convolutional networks and applications in vision, Proc IEEE International Symposium on Circuits and Systems,2010, 253–256.
  • Blei D M, Ng A Y, Jordan M I, Latent dirichlet allocation, J Machine Learn Res, 3 (2003) 993–1022.
  • Negrel R, Picard D, Gosselin P H, Evaluation of second-order visual features for land-use classification, in International Workshop on Content-Based Multimedia Indexing, 2014, 1–5.
  • Lazebnik S, Schmid C, Ponce J, Beyond bags of features: Spatial pyramid matching for recognizing natural scene categories, Proc IEEE Conference on Computer Vision and Pattern Recognition, 2, 2006, 2169–2178.
  • Perronnin F, Sanchez J, Mensink T, Improving the fisher kernel for large-scale image classication, Proc European Conference on Computer Vision, 2010, 143–156.

Abstract Views: 9

PDF Views: 2




  • Scene based Classification of Aerial Images using Convolution Neural Networks

Abstract Views: 9  |  PDF Views: 2

Authors

Palak Mahajan
Department of Computer Science & IT, University of Jammu, J&K, India
Pawanesh Abrol
Department of Computer Science & IT, University of Jammu, J&K, India
Parveen K. Lehana
Department of Electronics, University of Jammu, J&K, India

Abstract


The advent of computer vision and evolution of high-end computing in remote sensing images have embellish various researchers for unprecedented development in remotely sensed aerial images. The requirement to extract essential information stimulated anatomization of aerial images for its usefulness. Deep learning provides state of the art solutions for widely explored visual recognition system and has emerged as an evolutionary area, being applicable to large scale image processing applications. Convolutional Neural Networks (CNNs), an essential component of deep learning algorithms consists of increasing the depth and connections in the processing layers to learn various features of data at different abstract levels. In this paper, we present an outlook for classifying and extracting the features of aerial images using CNN. We propose a CNN architecture based on various parameters and layers for classification. CNN has been evaluated on two publicly available aerial data sets: UC Merced Land Use and RSSCN7. Experimental results show that the proposed CNN architecture is competent and efficient in terms of accuracy as performance evaluation parameter in comparison with conventional classifiers like Bag of Visual Words (BOVW).

Keywords


CNN, Deep Learning, Feature Extraction, Image Classification.

References