Abstract
Classification of remotely sensed data requires the modelling of suitable image processing algorithms. The rise of machine learning systems upgraded the viability of satellite image applications. Using Convolutional Neural Networks (CNN), benchmark classification exactness can be accomplished for land cover grouping. Motivated by the concept of Normalized Difference Vegetation Index (NDVI), this paper utilizes only the red and near infrared (NIR) band information for classifying the publicly available SAT-4 and SAT-6 datasets. This is done, since NDVI computation requires only the two band (red and NIR) information and the classes involved in the dataset are types of vegetation. In this work, new deep learning architectures for three different networks (AlexNet, ConvNet, VGG) were proposed by hypertuning the network and the input as two band data. The modified architectures with the two band information along with reduced number of filters were trained and tested model manages to classify the images into different classes. The proposed architectures are compared against the existing architectures in terms of accuracy, precision and trainable parameters. The proposed architecture is found to perform equally efficient by retaining high accuracy with less number of trainable parameters, when compared against the the performance of benchmark deep learning architectures for satellite image classification.
Similar content being viewed by others
References
Audebert N, Le Saux B, Lefèvre S (2017) Segment-before-detect: vehicle detection and classification through semantic segmentation of aerial images. Remote Sens 9(4):368
Basu S, Ganguly S, Mukhopadhyay S, DiBiano R, Karki M, Nemani R (2015) Deepsat: a learning framework for satellite imagery. In: Proceedings of the 23rd SIGSPATIAL international conference on advances in geographic information systems, p 37
Bragilevsky L, Bajić IV (2017) Deep learning for Amazon satellite image analysis. Commun Comput Signal Process (PACRIM), 1–5
Chen H, Wang Y, Gao S (2017) Assessing relationship of air quality index and vegetation type using hyperspectral remote sensing. In: Geoscience and remote sensing symposium (IGARSS), pp 4878–4881
Chippy J, Jacob NV, Renu RK, Sowmya V, Soman K (2017) Least square denoising in spectral domain for hyperspectral images. Procedia Comput Sci 115:399–406
Dahigamuwa T, Yu Q, Gunaratne M (2016) Feasibility study of land cover classification based on normalized difference vegetation index for landslide risk assessment. Geosciences 6(4):45
Dev S, Wen B, Lee YH, Winkler S (2016) Ground-based image analysis: a tutorial on machine-learning techniques and applications. IEEE Geosci Remote Sens Mag, 79–93
Dixon K, Deepa M, Ajay A, Sowmya V, Soman KP (2016) Aerial and satellite image denoising using least square weighted regularization method. Ind J Sci Technol 9(30):1–10
Dutta S, Manideep BC, Rai S, Vijayarajan V (2017) A comparative study of deep learning models for medical image classification. IOP Conf Series: Mater Sci Eng 263(4):042097
Haridas N, Aswathy C, Sowmya V, Soman K (2016) Hyperspectral image denoising using legendre-fenchel transform for improved sparsity based classification. Intell Syst Technol Appl, 521–528
Jeevalakshmi D, Reddy SN, Manikiam B (2016) Land cover classification based on NDVI using LANDSAT8 time series: a case study Tirupati region. Commun Signal Process (ICCSP), 1332–1335
Kaiser P, Wegner JD, Lucchi A, Jaggi M, Hofmann T, Schindler K (2017) Learning aerial image segmentation from online maps. IEEE Trans Geosci Remote Sens 55(11):6054–6068
Li H, Tao C, Wu Z, Chen J, Gong J, Deng M (2017) RSI-CB: a large scale remote sensing image classification benchmark via crowdsource data. arXiv:1705.10450
Liu Q, Hang R, Song H, Li Z (2018) Learning multiscale deep features for high-resolution satellite image scene classification. IEEE Trans Geosci Remote Sens 56 (1):117–26
Lu D, Weng Q (2007) A survey of image classification methods and techniques for improving classification performance. Int J Remote Sens 28(5):823–870
Lunga D, Yang HL, Reith A, Weaver J, Yuan J, Bhaduri B (2018) Domain-adapted convolutional networks for satellite image classification: a large-scale interactive learning workflow. IEEE J Selected Topics Appl Earth Observ Remote Sens 11(3):962–77
Makantasis K, Karantzalos K, Doulamis A, Doulamis N (2015) Deep supervised learning for hyperspectral data classification through convolutional neural networks. Geosci Remote Sens Symp (IGARSS), 4959–4962
Moorthi SM, Misra I, Kaur R, Darji NP, Ramakrishnan R (2011) Kernel based learning approach for satellite image classification using support vector machine. Recent Adv Intell Comput Syst (RAICS), 107–110
Nath SS, Mishra G, Kar J, Chakraborty S, Dey N (2014) A survey of image classification methods and techniques. Control, Instrum, Commun Comput Technol (ICCICCT), 554–557
Nie L, et al (2017) Enhancing micro-video understanding by harnessing external sounds. ACM Int conf on multimedia, pp 1192–1200
Özbay B, Ċimtay Y, Kandaz F (2017) Calculation of vegetation index for short wave infrared hyperspectral images. In: Signal processing and communications applications conference (SIU), pp 1–3
Paisitkriangkrai S, Sherrah J, Janney P, van den Hengel A (2016) Semantic labeling of aerial and satellite imagery. IEEE J Selected Topics Appl Earth Observ Remote Sens 9(7):2868–2881
Papadomanolaki M, Vakalopoulou M, Zagoruyko S, Karantzalos K (2016) Benchmarking deep learning frameworks for the classification of very high resolution satellite multispectral data. ISPRS Ann Photogram Remote Sens Spat Inf Sci 3(7):83–88
Sachin Rajan, Sowmya V, Govind D, Soman KP (2017) Dependency of various color and intensity planes on CNN based image classification. In: International symposium on signal processing and intelligent recognition systems, pp 167–177
Song S, et al (2017) NeuroStylist: neural compatibility modeling for clothing matching. ACM Int conf on multimedia, pp 753–761
Song S, et al (2018) Neural compatibility modeling with attentive knowledge distillation. In: Int ACM SIGIR conference on research & development in information retrieval, pp 5–14
Srivatsa S, Sowmya V, Soman K (2016) Least square based fast denoising approach to hyperspectral imagery. Intell Comput Techniques: Theory Practice Appl, 22–24
Xu D, Sun L, Luo J, Liu Z (2013) Analysis and denoising of hyperspectral remote sensing image in the curvelet domain. Mathematical Problems in Engineering
Zhang C, Hao X, Bai J, Dai M (2014) Improving hyperspectral data classification of satellite imagery by using a sparse based new model with learning dictionary. Hyperspectral Image Signal Process Evol Remote Sens, 1–4
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Unnikrishnan, A., Sowmya, V. & Soman, K.P. Deep learning architectures for land cover classification using red and near-infrared satellite images. Multimed Tools Appl 78, 18379–18394 (2019). https://doi.org/10.1007/s11042-019-7179-2
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-019-7179-2