Assessment of Multi-Sensor Neural Image Fusion and Fused Data Mining for Land Cover Classification

Recent studies suggest that the combination of imagery from earth observation satellites with complementary spectral, spatial, and temporal information may provide improved land cover classification performance. This paper assesses the benefits of new biologically-based image fusion and fused data mining methods for improving discrimination between spectrally-similar land cover classes using multi-spectral, multi-sensor, and multi-temporal imagery. For this investigation multi-season Landsat and Radarsat imagery of a forest region in central New York State was processed using opponent-band image fusion, multi-scale visual texture and contour enhancement, and the fuzzy ARTMAP neural classifier. These methods are shown to enable identification of sub-categories of land cover and provide improved classification accuracy compared to traditional statistical methods

1 Introduction

During the past decade significant advances in satellite sensor technology, data processing techniques, and computational power have made it possible to obtain new information about earth features and their relationships from a more global perspective. Recent advances in multi-sensor fusion are making it possible to combine complementary information from multiple regions of the electromagnetic spectrum. Several studies suggest that the combination of remotely-sensed electro-optical and microwave data may improve overall land cover classification accuracy [1]. Fusion of visible and infrared electro-optical data and synthetic aperture radar data, from different regions of the electromagnetic spectrum, may provide increased interpretation capabilities for forest classification [2] [3] [4]. Temporal data which includes seasonal changes can be used to add another dimension of useful information for classification [5].