Robust industrial defect identification, classification and quantification using low resolution non-destructive inspection techniques for automated digital shadowing

The Doctoral Candidate will be expected to develop Machine Learning tools that enable automated, objective and efficient identification, classification and quantification (ICQ) and spatial mapping of meso- and macro-scale defects in composite materials, through training on coupled high-resolution and low-resolution non-destructive inspection data (XRM and industrial ultrasonic imaging data). The successful candidate will develop scientific concepts and communicate research results through scientific publications and presentations at international conferences. The candidate will collaborate closely with fellow doctoral candidates within the LEGEND network, exploiting synergies across projects, and will actively participate in General Assembly meetings, training events, and international secondments across Europe. The research will focus on improving industrial non-destructive inspection techniques, developing deep-learning based tools for automated defect characterisation, and contri...