Research Papers

Classification of soybean varieties using different techniques: case study with Hyperion and sensor spectral resolution simulations

[+] Author Affiliations
Fábio M. Breunig

Instituto Nacional de Pesquisas Espaciais (INPE), Divisão de Sensoriamento Remoto, Caixa Postal 515, 12245-970, São José dos Campos, São Paulo, Brazil

Lênio S. Galvão

Instituto Nacional de Pesquisas Espaciais (INPE), Divisão de Sensoriamento Remoto, Caixa Postal 515, 12245-970, São José dos Campos, São Paulo, Brazil

Antônio R. Formaggio

Instituto Nacional de Pesquisas Espaciais (INPE), Divisão de Sensoriamento Remoto, Caixa Postal 515, 12245-970, São José dos Campos, São Paulo, Brazil

José C. N. Epiphanio

Instituto Nacional de Pesquisas Espaciais (INPE), Divisão de Sensoriamento Remoto, Caixa Postal 515, 12245-970, São José dos Campos, São Paulo, Brazil

J. Appl. Remote Sens. 5(1), 053533 (June 29, 2011). doi:10.1117/1.3604787
History: Received December 02, 2010; Revised May 11, 2011; Accepted June 07, 2011; Published June 29, 2011; Online June 29, 2011
Text Size: A A A

Next generation imaging spectrometers with higher signal-to-noise ratio and broader swath-width bring new perspectives for crop classification over large areas. Here, we used Hyperion/Earth Observing-One data collected over Brazilian soybean fields to evaluate the performance of four classification techniques (maximum likelihood — ML; spectral angle mapper — SAM; spectral information divergence — SID; support vector machine — SVM) to discriminate five soybean varieties. The spectral resolution influence on classifying them was analyzed by simulating the spectral bands of seven multispectral sensors using Hyperion data. Before classification, the Waikato environment for knowledge analysis was used for feature selection. Results showed the importance of the green, red-edge, near-infrared, and shortwave infrared to discriminate the soybean varieties. Because the soybean variety Monsoy 8411 was sensed by Hyperion in a later reproductive stage, it was more easily discriminated than the other varieties. The best classification techniques were ML and SVM with overall accuracy of 89.80% and 81.76%, respectively. The accuracy of spectral matching techniques was lower (70.84% for SAM and 72.20% for SID). When ML was applied to the simulated spectral resolution of the multispectral sensors, moderate resolution imaging spectroradiometer and enhanced thematic mapper plus presented the highest accuracy, whereas advanced very high resolution radiometer showed the lowest one.

Figures in this Article
© 2011 Society of Photo-Optical Instrumentation Engineers (SPIE)

Citation

Fábio M. Breunig ; Lênio S. Galvão ; Antônio R. Formaggio and José C. N. Epiphanio
"Classification of soybean varieties using different techniques: case study with Hyperion and sensor spectral resolution simulations", J. Appl. Remote Sens. 5(1), 053533 (June 29, 2011). ; http://dx.doi.org/10.1117/1.3604787


Access This Article
Sign in or Create a personal account to Buy this article ($20 for members, $25 for non-members).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Related Book Chapters

Topic Collections

Advertisement
  • Don't have an account?
  • Subscribe to the SPIE Digital Library
  • Create a FREE account to sign up for Digital Library content alerts and gain access to institutional subscriptions remotely.
Access This Article
Sign in or Create a personal account to Buy this article ($20 for members, $25 for non-members).
Access This Proceeding
Sign in or Create a personal account to Buy this article ($15 for members, $18 for non-members).
Access This Chapter

Access to SPIE eBooks is limited to subscribing institutions and is not available as part of a personal subscription. Print or electronic versions of individual SPIE books may be purchased via SPIE.org.