The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences (Jan 2014)

MULTI-TEMPORAL LAND USE ANALYSIS OF AN EPHEMERAL RIVER AREA USING AN ARTIFICIAL NEURAL NETWORK APPROACH ON LANDSAT IMAGERY

  • M. Aquilino,
  • E. Tarantino,
  • U. Fratino

DOI
https://doi.org/10.5194/isprsarchives-XL-5-W3-167-2013
Journal volume & issue
Vol. XL-5/W3
pp. 167 – 173

Abstract

Read online

This paper proposes a change detection analysis method based on multitemporal LANDSAT satellite data, presenting a study performed on the Lama San Giorgio (Bari, Italy) river basin area. Based on its geological and hydrological characteristics, as well as on the number of recent and remote flooding events already occurred, this area seems to be naturally prone to flooding. The historical archive of LANDSAT imagery dating back to the launch of ERTS in 1972 provides a comprehensive and permanent data source for tracking change on the planet‟s land surface. In this study case the imagery acquisition dates of 1987, 2002 and 2011 were selected to cover a time trend of 24 years. Land cover categories were based on classes outlined by the Curve Number method with the aim of characterizing land use according to the level of surface imperviousness. After comparing two land use classification methods, i.e. Maximum Likelihood Classifier (MLC) and Multi-Layer Perceptron (MLP) neural network, the Artificial Neural Networks (ANN) approach was found the best reliable and efficient method in the absence of ground reference data. The ANN approach has a distinct advantage over statistical classification methods in that it is non-parametric and requires little or no a priori knowledge on the distribution model of input data. The results quantify land cover change patterns in the river basin area under study and demonstrate the potential of multitemporal LANDSAT data to provide an accurate and cost-effective means to map and analyse land cover changes over time that can be used as input in land management and policy decision-making.