SUMMARY
Visual inspections comprise the majority of inspections for large transport aircraft and are traditionally conducted by human operators. The manual inspection process is time consuming, inconsistent, and subject to human errors. Automated defect detection systems have been developed to leverage computer vision and deep learning to decrease inspection times and improve detection performance. Current state-of-the-art systems use convolutional neural networks (CNNs) to detect defects from image data. The performance of these systems is insufficient for critical aircraft inspection and there is little consideration for the balance of false alarms and missed detections. This thesis presents a novel application of deep learning ensembles to automated aircraft visual inspection to improve the performance of CNNs and provide a framework for managing the tradeoff between Type I and Type II error. The performance of the stacked ensembles is evaluated, and it is found that stacked ensembles of CNNs outperform the current state-of-the-art defect detection approaches.Meeting URLhttps://bluejeans.com/522575092 Meeting ID522 575 092Want to dial in from a phone?Dial one of the following numbers:+1.888.748.9073 (United States(Primary))+1.844.540.8065 (United States(Primary))+1.408.419.1715 (United States(San Jose))+1.408.915.6290 (United States(San Jose))(see all numbers - https://www.bluejeans.com/premium-numbers)Enter the meeting ID and passcode followed by #Connecting from a room system?Dial: bjn.vc or 199.48.152.152 and enter your meeting ID & passcode