Contextual land-cover classification: incorporating spatial dependence in land-cover classification models using random forests and the Getis statistic

Land-cover characterization of large heterogeneous landscapes is challenging because of the confusion caused by high intra-class variability and heterogeneous landscape artefacts. Neighbourhood context can be used to supplement spectral information, and a novel way of incorporating spatial dependence in a heterogeneous region is tested here using an ensemble learning technique called random forests and a measure of local spatial dependence called the Getis statistic. The overall Kappa accuracy of the random forest classifier that used a combination of spectral and local spatial (Getis) variables at three different neighbourhood sizes (3 × 3, 7 × 7, and 11 × 11) ranged from 0.85 to 0.92. This accuracy was higher than that of a non-spatial random forest classifier having an overall Kappa accuracy of 0.78, which was run using the spectral variables only. This study demonstrated that the use of the Getis statistic with different neighbourhood sizes leads to substantial increase in per class classification accuracy of heterogeneous land-cover categories.

[1]  J Richards,et al.  Computer processing of remotely-sensed images: An introduction , 1990 .

[2]  Russell G. Congalton,et al.  A review of assessing the accuracy of classifications of remotely sensed data , 1991 .

[3]  Stephen V. Stehman,et al.  Selecting and interpreting measures of thematic classification accuracy , 1997 .

[4]  Michael A. Wulder,et al.  Local spatial autocorrelation characteristics of remotely sensed imagery assessed with the Getis statistic , 1998 .

[5]  John A. Richards,et al.  Remote Sensing Digital Image Analysis: An Introduction , 1999 .

[6]  Philip Lewis,et al.  Geostatistical classification for remote sensing: an introduction , 2000 .

[7]  Marvin E. Bauer,et al.  Integrating Contextual Information with per-Pixel Classification for Improved Land Cover Classification , 2000 .

[8]  Michael A. Wulder,et al.  Remote sensing methods in medium spatial resolution satellite data land cover classification of large areas , 2002 .

[9]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[10]  Thomas Blaschke,et al.  Image Segmentation Methods for Object-based Analysis and Classification , 2004 .

[11]  Joydeep Ghosh,et al.  Investigation of the random forest framework for classification of hyperspectral data , 2005, IEEE Transactions on Geoscience and Remote Sensing.

[12]  Panayiotis E. Pintelas,et al.  Combining Bagging and Boosting , 2007 .

[13]  Mahesh Pal,et al.  Random forest classifier for remote sensing classification , 2005 .

[14]  John Rogan,et al.  6 Integrating GIS and Remotely Sensed Data for Mapping Forest Disturbance and Change , 2006 .

[15]  Steven E. Franklin,et al.  Understanding Forest Disturbance and Spatial Pattern : Remote Sensing and GIS Approaches , 2006 .

[16]  Johannes R. Sveinsson,et al.  Random Forests for land cover classification , 2006, Pattern Recognit. Lett..

[17]  Soe W. Myint,et al.  Employing Spatial Metrics in Urban Land-use/Land-cover Mapping: Comparing the Getis and Geary Indices AAG REMOTE SENSING SPECIALITY GROUP 2007 AWARD WINNER 1 , 2007 .

[18]  C. Lippitt,et al.  Mapping Selective Logging in Mixed Deciduous Forest: A Comparison of Machine Learning Algorithms , 2008 .

[19]  Janet Franklin,et al.  Mapping land-cover modifications over large areas: A comparison of machine learning algorithms , 2008 .

[20]  A. Getis.  The Analysis of Spatial Association by Use of Distance Statistics , 2010 .

[21]  J. Ord,et al.  Local Spatial Autocorrelation Statistics: Distributional Issues and an Application , 2010 .