Measuring subsurface defect depth and metal loss by thermal wave imaging and inverse scattering of photon density waves

Since late 1980s, thermal wave imaging has becoming a more and more powerful technique for nondestructive testing (NDT), sometimes called "nondestructive evaluation" (NDE) due to its unique characteristics: fast, non-contact and non-invasive, covers wide areas, and operates easily with single sided access to the item being inspected. This new NDT technique has been used to identify subsurface defects in various samples which are made from materials with different thermal properties, such as metals, alloys, plastics, ceramics, and composites. Several years ago we started to study the early-time behavior of thermal waves and work on the quantitative measurement of subsurface defects. As a result, we found a useful parameter for measuring the absolute depth of subsurface defects, regardless of their lateral sizes. Under the driving force of measuring corrosion thinning on aircraft, we developed another method to measure the relative thickness of good thermal conductors like aluminum. The resulting algorithms from both of the methods have been put into a thermal wave imaging software package and have been successfully applied to real-world NDE. Thermal wave imaging systems have been taken to the FAA-Center for Aviation System Reliability at Sandia, Boeing, Northwest Airlines, etc., for detecting defects and estimation of corrosion metal loss. The theory associated with these two methods are given in this dissertation, so are the theoretical and experimental results, as well as the comparison. As the second part of this dissertation, inverse scattering of photon density waves is described. Image reconstruction has always been a very important area of research. It has been used in many different fields. We all know that due to thermal diffusion, thermal wave images of deep subsurface structures get blurred. To reconstruct the subsurface structures, an inverse scattering algorithm was developed several years ago, and was successfully used in recovering scatterer shapes from thermal wave images. Similarly, due to photon diffusion, the optical image of an object in a turbid liquid, say milk, also gets blurred. To reconstruct the shape of an optical scatterer in a turbid medium, we developed an inverse scattering algorithm which deals with inversion of photon density wave images. The resulting inversion technique is capable of making a significant improvement in image quality through reconstruction of an object. The inverted image basically recovers the shape and the size of the object. Profile plots show that the resulting image has a flat top in contrast to the original rounded top of the blurred image. We conclude that the inversion algorithm works very well.