Speaker: Sergio Segovia González.

Abstract: New way to automatize the inspection of nuclear fuel elements in spent fuel storage processing video signal and audio signal. For video signal, it is developed a custom database including images from several nuclear facilities in Spain, fine-tuned pre-trained object detection systems to recognize the characters and built a system to recognize the complete identification codes of fuel heads. For audio signal, it is development an algorithm that uses as a basis one of the latest models of multilingual Automatic Speech Recognition to transcribe audio signal, and with a post-process of the transcriptions segments we build the identification code of fuel head and other components. Results show a very high accuracy in images and even more in videos, where several frames provide more robustness than a single image addition the results in audios from these videos show a very high accuracy too. Besides, the methodology proposed is easily extensible to other nuclear facilities in the world and even to different alphanumeric characters if required.