Classification of Field-Level Crop Types with a Time Series Satellite Data Using Deep Neural Network

  • J. Jayanth
  • V. S. Shalini
  • T. Ashok Kumar
  • Shivaprakash Koliwad
Part of the Remote Sensing and Digital Image Processing book series (RDIP, volume 24)


Crop-type classification has been relied upon on only spectral/spatial features. It does not provide the in-season information for researchers and decision makers for both practical and scientific purposes. While satellite images have desirable spectral and spatial information for classification, the ability to extract temporal information in satellite data remains a challenge due to revisiting frequency and gaps in the time period of capturing the data. To circumvent this challenge and generate more accurate results for an in-season crop-type classification, we have used Rectified Linear Unit (RLU) approach based on the concept of deep neural networks for intelligent and scalable computation of the classification process. The work was carried out on Nanjangud Taluk located in Mysuru District, Karnataka state on a Landsat data (multi-temporal scene) from 2010 to 2015. The results indicate that RLU shows an improvement of 5% to 15% for overall classification accuracy at 3 classes over the traditional against support vector machine. In comparison with KSRSC data set, this study reveals an accuracy of 85% for classifying rice and banana with an improvement of 10% over KSRCS crop-filed data.


Spectral Temporal Landsat RLU Rice Banana 



The author graciously thanks Dr. Dwarkish G S, professor, Hydraulics Department, NITK, Mangalore, for providing the remote-sensed data for this study.


  1. 1.
    Bolton DK, Friedl MA (2013) Forecasting crop yield using remotely sensed vegetation indices and crop phenology metrics. Agric For Meteorol 173:74–84. CrossRefGoogle Scholar
  2. 2.
    Gao F, Anderson MC, Zhang X, Yang Z, Alfieri JG, Kustas WP, Mueller R, Johnson DM, Prueger JH (2017) Toward mapping crop progress at field scales through fusion of Landsat and MODIS imagery. Remote Sens Environ 188:9–25. CrossRefGoogle Scholar
  3. 3.
    King L, Adusei B, Stehman SV, Potapov PV, Song X-P, Krylov A, Di Bella C, Loveland TR, Johnson DM, Hansen MC (2017) A multi-resolution approach to national-scale cultivated area estimation of soybean. Remote Sens Environ 195:13–29. CrossRefGoogle Scholar
  4. 4.
    Hinton G, Deng L, Yu D, Dahl G, Mohamed A, Jaitly N, Senior A, Vanhoucke V, Nguyen P, Sainath T, Kingsbury B (2012) Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups. IEEE Signal Process Mag 29(6):82–97. CrossRefGoogle Scholar
  5. 5.
    Krizhevsky A, Sutskever I, Hinton GE (2012). ImageNet Classification With Deep Convolutional Neural Networks. Advances in Neural Information Processing SystemsGoogle Scholar
  6. 6.
    LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521(7553):436–444. CrossRefGoogle Scholar
  7. 7.
    Schmidhuber J (2015) Deep learning in neural networks: an overview. Neural Netw 61:85–117. CrossRefGoogle Scholar
  8. 8.
    Schmidt G, Jenkerson C, Masek J, Vermote E, Gao F (2013) Landsat Ecosystem Disturbance Adaptive Processing System (LEDAPS) Algorithm Description. US Geological SurveyGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • J. Jayanth
    • 1
  • V. S. Shalini
    • 2
  • T. Ashok Kumar
    • 3
  • Shivaprakash Koliwad
    • 4
  1. 1.Department of Electronics and Communication EngineeringGSSS Institute of Engineering & Technology for WomenMysoreIndia
  2. 2.Department of Electronics and Communication EngineeringATME College of EngineeringMysoreIndia
  3. 3.Sri Dharmasthala Manjunatheshwara Institute of TechnologyUjireIndia
  4. 4.Department of Electronics & Communication EngineeringMalnad College of EngineeringHassanIndia

Personalised recommendations