Segmentation scale parameter influence on the accuracy of detecting illegal landfills on satellite imagery. A case study for Novo Sarajevo

Abstract

Illegal waste dumping has been widely regarded as one of the biggest source of environmental damage. Illegal landfills are a prevailing problem existing in a large number of countries. To control and better manage illegal landfills, it is necessary to know the current locations and contents of illegal landfills. This could increase efficiency in illegal landfill management and preserve the biodiversity and ecological balance. Remote sensing methods have been proven extremely effective in detecting potential illegal landfill sites. This paper investigates the relationship between the segmentation scale parameter and the detection accuracy of illegal landfill sites in urban areas that are not covered by vegetation or buried in the ground. The research showed that there is an optimal scale parameter (SP = 20) value for the used satellite image Pléiades 1B and area of interest (Novo Sarajevo municipality). The scale parameter's stated value gives maximum Kappa values and Overall accuracy coefficients for detected illegal landfills on the satellite image.