Quantum machine learning ensemble for surface crack detection

Abstract

By identifying the aspects of manual inspection methods in the context of industrial production, which are described within the undertaken research, the development of an automated visual inspection technology is driven. This causes more time to be spent on performing the checks, thus adding to the labor cost. The efficiency of the operations is reduced, and there is a tendency for errors due to fatigue in checking 24/7. The proposed solution for a new product is designed to change the approach of the existing manufacturing process by using the automated system to self-inspect the surface and notify of its defects during manufacturing. As an enhancing advancement, this new development aims to address apprehensions pertaining to manual examination as the world transitions into the fault tolerant period. Lastly, this approach fits the universal grail of further developing industrial capacities, with the resulting thought process extending to the incorporation of technologies such as quantum computing with the current requirements of manufacturing. Other potential applications of this approach, including aerospace applications of ultrasonic testing or thermography in the detection of surface cracks, might also help improve this approach in the future.