Urban Origins/Destinations from High-Resolution Remote Sensing Images

Abstract

High-resolution remote sensing (RS) image data were used to identify commercial and industrial (C&I) origins and destinations (O/D). Imperviousness, derived from a RS-based land cover classification, is utilized as a surrogate for C&I locations. Imperviousness is quantified using three parameters that indicate the percent of impervious surface in a block of interest and in surrounding blocks, each quantified through an intensity of red, green, or blue. The three parameters are combined in a meaningful way through the combination of the intensity of the three colors to represent which impervious surfaces are associated with C&I locations. A block size analysis was performed to determine the block size that best differentiates between C&I and non-C&I locations. Training sites were used to develop a land cover classification with two classes—C&I or non-C&I, that incorporates the color variations associated with C&I locations, including the impact of boundaries. An accuracy assessment was performed by comparing C&I/non-C&I designations with actual land use. C&I O/D are of use in determining travel distances and as a measure of transportation system accessibility.