Our project introduces an innovative cloud-based AI application designed for engine borescope digital inspections in the aerospace industry. The aim being to use artificial intelligence, more specifically computer vision, to streamline the inspection process by detecting and specifying defects of aircraft engines.
Traditional engine borescope inspections have many limitations, they are not only complex, and time-consuming but demand the manual labor of highly specialized personnel. The process is also prone to human error, most commonly fatigue which can often result in a failure to upkeep quality inspections.
Recognizing these challenges, our project sought to address the downsides of manual inspections by leveraging computer vision to enhance the precision, speed, and constancy of the inspection process. The katana application acted as a cloud-based assistant, supporting inspectors during the borescope inspections. This was done so by the AI system which employs computer vision algorithms to detect and specify defects, precisely determine their positions, and account for complete blade coverage. By automating these tasks, the burden on inspectors was significantly reduced, allowing them to focus on high-level analysis.
Although a very new project, conducted purely for research purposes the implementation of the Katana application gave way to numerous benefits, most importantly supporting inspectors, leading to higher inspection quality and increased customer satisfaction. The collaborative aspect of the project allows experts from around the world to review borescope results, promoting knowledge sharing and in turn leading to continuous improvement.
Furthermore, the collected inspection data serves as a valuable resource for training new inspectors and contributes to building up a history of defects, enabling predictive maintenance strategies.
