Author(s): Tahar El-Korchi, Michael A. Gennert, Matthew O. Ward, and Norman Wittels, Worcester Polytechnic Institute, Worcester, MA 01609

Source: Proc. Automated Pavement Distress Data Collection Equipment Seminar, pp165-174, Iowa State University, Ames IA, 12-15 June 1990.

Abstract: Automated pavement surface distress evaluation is economically important but has exhibited slow growth because the accuracy of automated systems has not been sufficient to inspire confidence among the pavement engineers who have to rely on the evaluation results. System performance would be enhanced by using better engineering methods and data in designing the image acquisition and image processing portions of automated inspection systems.

In this paper we outline an engineering approach comprising two steps: first, a study of the signal characteristics of pavement video images - the apparent luminance of cracks in pavements and the contrast that they exhibit compared to surrounding pavement surfaces - and, second, the design of inspection systems that are optimized to those characteristics. We then describe results of our studies of image luminance. We summarize how the crack luminance depends on the reflectivities of paving materials and we show how they can be used with computer modeling in designing automated pavement inspection systems. As an example, we show, under some conditions, that ambient lighting (such as on an overcast day) produces fewer missed cracks than other forms of lighting.