<title>Automated inspection system for detecting metal surface cracks from fluorescent penetrant images</title>

Abstract

Cracks occurred in aircraft engine parts have to be detected as early as possible to prevent engine failure. Fluorescent Penetrant Inspection (FPI), that applies fluorescent materials on metallic surfaces for flaw detection, is a generally accepted technology for nondestructive inspection of surface cracks. The major problem with application of FPI technology is the costly false alarms caused by non-crack fluorescence indications (noise), especially when inspecting used engine parts.A novel crack-detection system for automatic FPI of engine parts using image processing and pattern recognition theories is presented. A strong noise reduction capability and a small number of reliable features for pattern recognition are the two primary characteristics of the system, which contains three major modules: noise-reduction and preclassifier module, feature extraction module, and pattern recognition module including four pattern classifiers.An image synthesizing technique is developed to simulate real-world situations by combining the segmented fluorescence images of man-made cracks with the noisy background of fluorescent images captured from actual used parts. The designed system can eliminate over 80% of noise while retain 94% of crack indication. The total error rate using Fisher's linear classifier is less than 3%, with only 4% of crack misclassification.

Using ResNets to perform automated defect detection for Fluorescent Penetrant Inspection

“…Previous work looking into automated defect detection for FPI has investigated the use of traditional image processing techniques [37,5,6,41,21] and, more recently, using the Random Forest machine learning method [36,35]. However, image processing methods are often too reliant on a specic shape of defect, small amounts of background penetrant, defects being above a certain size etc.…”

Defect Detection and Classification by Training a Generic Convolutional Neural Network Encoder

“…N ON-DESTRUCTIVE Testing (NDT) [1], [2] is widely used in quality control and monitoring of manufactured components. Different techniques can be used for NDT purposes, including ultrasonic testing (UT) [3], radiographic testing (RT) [4], [5], penetrant testing (PT) [6] and so on. Traditional NDT techniques are normally performed by human inspectors.…”

Automated defect detection for Fluorescent Penetrant Inspection using Random Forest

“…The rst attempt at this was work by Tang in 1995 [35]. A threshold method based on boundary shapes was used, followed by pattern classication methods.…”

“…A threshold method based on boundary shapes was used, followed by pattern classication methods. Articially induced cracks which were superimposed onto images of engine run components were detected with a correct call rate of 97.7% [35].…”