Noncontact laser metrology with real-time detection and high-speed processing for material analysis

Abstract

This paper describes the development of real-time non-contact metrology based on holographic interferometry and neural network fringe analysis software to detect defects in composite materials. The object under inspection is illuminated by a high-power solid-state laser, and the light scattered from the material surface is recorded in real time using a double-exposure holographic interferometer, which can detect minute surface deformation caused by defects in the material. Thermal stimulation of the object creates a non- uniform time-varying material reaction, which causes surface deformation that is characteristic of the material's internal structure. This deformation in turn creates time-varying interference patterns, which are recorded by a real-time holographic interferometric system and displayed on the computer monitor through a CCD camera. The system allows real- time, in-depth non-contact inspection of composite materials used in aircraft and other military vehicles. A genetic algorithm has also been developed for fast data processing in a non-laboratory environment. The sophisticated neural network recognizes the types of defects at high speed.