Non-destructive testing (NDT) of aircraft parts has become increasingly important in improving the safety and reliability of the aerospace industry, especially in the testing of high-temperature and high-pressure turbine engine parts. Among the various types of NDT methods available, fluorescent penetrant inspection (FPI) is comparably more cost-efficient and is widely used in NDT on aircraft parts. However, current FPI still requires considerable labour forces in its processing, inspection and analysis procedures. In this paper, we have developed an advanced automatic inspection system (AAIS) that uses image processing and pattern recognition techniques to aid human inspectors. The system can automatically detect, measure and classify discontinuities from the FPI images of aircraft parts. Tests have been performed on the sample images provided by our industrial partners to evaluate our developed AAIS. The test results demonstrate that the developed system has significantly improved the efficiency of FPI with satisfactory accuracy.
Insight: Non-Destructive Testing and Condition Monitoring57, no. 1: 18–34.