Automated estimation of occupancy measure on beaches using digital image processing

Abstract

In this paper we describe a methodology for occupancy measure on beaches based on digital image processing techniques. This metric constitute a measure of interest for quantification of factors relates to the quality of user experience and is an import indicator in integrated management field. The proposed method has two main components: the preprocessing and occupancy density estimation. In the preprocessing stage a color correction approach is applied in order to minimize the illumination variance effects over the objects in the scene. Subsequently, the estimation begins delimiting the information regarding the environment using a mask or reference image. Gray-level gradient analysis is applied to the resulted image after the mask appliance. The results show a 2.6% margin of error that suggests the measure is an occupancy measure indicative. The obtained error is related to the auto occlusion identification and processing.