Remote Sensing Applications and Decision Support

Integrating pan-sharpening and classifier ensemble techniques to map an invasive plant (Spartina alterniflora) in an estuarine wetland using Landsat 8 imagery

[+] Author Affiliations
Jinquan Ai, Runhe Shi, Chao Zhang, Chaoshun Liu

East China Normal University, College of Geographical Sciences, Shanghai 200241, China

East China Normal University, Key Laboratory of Geographic Information Science, Ministry of Education, Shanghai 200241, China

Wei Gao

East China Normal University, College of Geographical Sciences, Shanghai 200241, China

East China Normal University, Key Laboratory of Geographic Information Science, Ministry of Education, Shanghai 200241, China

Colorado State University, Natural Resource Ecology Laboratory, Fort Collins, Colorado 80523, United States

Zhiqiang Gao

Yantai Institute of Coastal Zone Research, Chinese Academy of Sciences, Yantai 264003, China

J. Appl. Remote Sens. 10(2), 026001 (Apr 07, 2016). doi:10.1117/1.JRS.10.026001
History: Received November 23, 2015; Accepted March 7, 2016
Text Size: A A A

Abstract.  Accurate mapping of invasive species in a cost-effective way is the first step toward understanding and predicting the impact of their invasions. However, it is challenging in coastal wetlands due to confounding effects of biodiversity and tidal effects on spectral reflectance. The aim of this work is to describe a method to improve the accuracy of mapping an invasive plant (Spartina alterniflora), which is based on integration of pan-sharpening and classifier ensemble techniques. A framework was designed to achieve this goal. Five candidate image fusion algorithms, including principal component analysis fusion algorithm, modified intensity-hue-saturation fusion algorithm, wavelet-transform fusion algorithm, Ehlers fusion algorithm, and Gram–Schmidt fusion algorithm, were applied to pan-sharpening Landsat 8 operational land imager (OLI) imagery. We assessed the five fusion algorithms with respect to spectral and spatial fidelity using visual inspection and quantitative quality indicators. The optimal fused image was selected for subsequent analysis. Then, three classifiers, namely, maximum likelihood, artificial neural network, and support vector machine, were employed to preclassify the fused and raw OLI 30-m band images. Final object-based S. alterniflora maps were generated through classifier ensemble analysis of outcomes from the three classifiers. The results showed that the introduced method obtained high classification accuracy, with an overall accuracy of 90.96% and balanced misclassification errors between S. alterniflora and its coexistent species. We recommend future research to adopt the proposed method for monitoring long-term or multiseasonal changes in land coverage of invasive wetland plants.

Figures in this Article
© 2016 Society of Photo-Optical Instrumentation Engineers

Citation

Jinquan Ai ; Wei Gao ; Zhiqiang Gao ; Runhe Shi ; Chao Zhang, et al.
"Integrating pan-sharpening and classifier ensemble techniques to map an invasive plant (Spartina alterniflora) in an estuarine wetland using Landsat 8 imagery", J. Appl. Remote Sens. 10(2), 026001 (Apr 07, 2016). ; http://dx.doi.org/10.1117/1.JRS.10.026001


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.