PHYSICALLY BASED ULTRASONIC FEATURE MAPPING FOR ANOMALY CLASSIFICATION IN COMPOSITE MATERIALS.

In the F-map process, detection of anomalies is performed by observing either a change in signal amplitude by traditional C-scan analysis, or by noting changes in the feature value of a backwall echo or in a through-transmission signal. Identification is then accomplished by examining features that are developed by using physical models that describe the interaction between anomaly geometry and ultrasonic wave propagation. Identification procedures using the ultrasonic F-map process are presented for three common anomaly types.