Active Contour Techniques for Automated Analysis of Digital X-Ray Images of Aircraft Health

An image processing methodology for the automatic analysis of digital X-Ray images of aircraft structures is proposed. The technical approach employs the Active Contours image segmentation algorithm. This algorithm utilizes a region, rather than edge based techniques. It has the advantage of being resistant to noise effects and being able to segment objects that do not have sharp edged contours. Furthermore, our technique is dependent on fewer tunable parameters than traditional approaches, hence can be easily extended to other frequency bands beyond X-Rays. This method will be particularly effective in detecting and classifying cracks and other anomalies in aircraft structures. The technology proposed has been previously applied to image processing in Missile Defense as well as medical applications. Our study will work with currently available x-ray imagery to identify key anomalies that need to be observed in order to develop a prototype that will be tested in the field.