Mapping the spatial performance variability of an X-ray computed tomography inspection

Lessons learned in the design of reference fiducials for layer-wise analysis of test coupons made by laser powder bed fusion

“…However, factors inherent to the X-ray CT instrument, such as detector unsharpness, the finite size of the X-ray source focal spot, and instrument drifts typically increase the minimum detectable feature. When measuring dense samples that succumb to significant X-ray scatter and beam hardening artifacts, it is expected that the minimum reliable feature volume will increase even further [43,44]. Task-specific probability of detection (POD) studies can provide a more accurate estimation of this minimum feature volume [45,46].…”

Review of the influence of noise in X-ray computed tomography measurement uncertainty

“…Reiter et al [90] studied both SNR and CNR for the different penetration lengths of an aluminium step cylinder, showing the dependence between travelled length and noise. The dependence between penetration length and noise level implies that, for this reason alone, noise will vary spatially across a volume [91], giving rise to the non-stationarity of noise (see section 2.2). Stationarity is normally used for the temporal domain; however, the term non-stationarity has been widely used in the literature to name the spatial variation of noise across XCT images [90,[92][93][94], and so is used here.…”

Advances in the UK Toward NDE 4.0

“…Past NDE research efforts worth highlighting given close links to the vision of NDE 4.0, without necessarily having been carried out in that context, include work on modelassisted probability of detection (MAPOD) [17], model-assisted inspection qualification [18][19][20], model-assisted data analysis [21][22], inspection optimization [23], data fusion [24], as well as inspection-based simulation [25][26][27] and machine leaning, even applied to a traditional, "wet" inspection technique [28]. Work on advanced robotic inspection and un-conventional inspection delivery systems, including collaborative robots (cobots) and unmanned aerial vehicles / drones, are worth highlighting, too, performed especially by the research and technology organization TWI and Strathclyde University [4,[29][30].…”

“…This MTC membership project, backed by a steering group of 20 companies, seeks to expand on previous efforts on MBD, specifically to digitalize part specifications relevant to NDE (and surface metrology / topology), incorporating them into MBD data, so as to bring these inspection domains in line with Industry 4.0 methodologies. The intention is then to demonstrate two applications of this approach: the streamlined co- (1) New NDE methods for new manufacturing: in AM, a shortfall in NDE capability, both at manufacture and in-service, has been identified as the primary barrier to structural use of AM components [20]. In intelligent process control, NDE information is required as an input, requiring new in-line inspection techniques that can be integrated into the build process [44].…”