The current method of assessing the condition of paint protective coatings is by human visual evaluation using ASTM pictorial guides for comparison and condition definition. This is purely subjective and the results are affected by area size and geometry and human interpretation. By developing an automated vision system that can enhance the paint surface image and provide an objective assessment, one would provide a means of obtaining more accurate and consistent results. The approach of this research is to use the ASTM pictorial guides and painted plate samples to establish the usual characteristics of the paint failure mechanisms. Using available electronic hardware, a number of programming techniques for image processing and analysis would be explored to provide assessment of enhanced images of paint. Both visual light and infrared red frequency band reflected energy imaging techniques will be included.
