Int. J. of Big Data Intelligence   »   2017 Vol.4, No.2

 

 

Title: Graph-based semi-supervised classification on very high resolution remote sensing images

 

Authors: Yupeng Yan; Manu Sethi; Anand Rangarajan; Ranga Raju Vatsavai; Sanjay Ranka

 

Addresses:
Department of Computer and Information Science and Engineering, University of Florida, Gainesville, FL 32611, USA
Department of Computer and Information Science and Engineering, University of Florida, Gainesville, FL 32611, USA
Department of Computer and Information Science and Engineering, University of Florida, Gainesville, FL 32611, USA
NC State University and Oak Ridge National Laboratory, 890 Oval Drive, Campus Box 8206, Raleigh, NC 27695, USA
Department of Computer and Information Science and Engineering, University of Florida, Gainesville, FL 32611, USA

 

Abstract: Classification of very high resolution (VHR) remote sensing imagery is a rapidly emerging discipline but faces several challenges owing to the huge scale of the pixel data involved, indiscernibility in the traditionally used features to represent various regions, and the lack of available ground truth data. This paper provides a framework which elegantly overcomes these hurdles by providing a novel semi-supervised learning approach which employs multiscale superpixel tessellation representations of VHR imagery. Superpixels are homogeneous and irregularly shaped regions which form the backbone of our approach and are used to derive novel features by learning a decision tree. Our semi-supervised learning approach works on a superpixel graph and seamlessly combines the large margin capability of a support vector machine (SVM) with a graph-based Laplacian label propagation approach to obtain a novel objective function. Further we also provide a self-contained and easily parallelisable linear iterative optimisation approach based on the principle of majorisation-minimisation. We evaluate this approach on four different geographic settings with varying neighbourhood types and draw comparisons with the popular and widely used Gaussian multiple instance learning algorithm. Our results showcase several advantages in accuracy and efficiency, which coupled with the ease of model building and inherently parallelisable optimisation make our framework a great choice for deployment in large scale applications like global human settlement mapping and population distribution, and change detection.

 

Keywords: image classification; high resolution images; remote sensing images; superpixel segmentation; image segmentation; ultrametric contour maps; majorisation-minimisation; support vector machines; SVM; graph Laplacian; semi-supervised learning; Gaussian multiple instance learning; label propagation; surrogate function.

 

DOI: 10.1504/IJBDI.2017.10002925

 

Int. J. of Big Data Intelligence, 2017 Vol.4, No.2, pp.108 - 122

 

Submission date: 10 Mar 2016
Date of acceptance: 12 Mar 2016
Available online: 01 Feb 2017

 

 

Editors Full text accessAccess for SubscribersPurchase this articleComment on this article