You are here

Advanced Machine Learning for Non-Destructive Testing



OBJECTIVE: Apply Machine Learning/Artificial Intelligence to aid in interpretation of radiography inspection results during non-destructive testing. 

DESCRIPTION: The Army relies on radiography inspection (e.g. x-ray and neutron) for non-destructive testing of munitions during production and special investigations. Interpreting the visual results of the inspections is a challenge and requires highly trained individuals (Level III Radiographers) to determine what, if any, problems actually exist. This topic will apply advanced and innovative machine learning and/or artificial intelligence to current and future non-destructive radiography inspection methods that use electronic imaging to identify defects and aid the operator in proper and timely interpretation of the results. As this technology is meant to be incorporated in a production line, the expectation is that it will support three dimensional inspection and interpretation of defects at a production rate of up to 1 unit per minute, and items up to 6.5 inches in diameter. Defects include cavities, porosity, piping, voids, gaps, low density, annular rings, cracks and inclusions ranging from 0.002" to 0.020". The technology must reside on a standard computer system linked to the inspection equipment and receive the electronic images from the radiography system. Specific interface requirements will be provided after contract award. This topic will also develop and deliver the output screens that provide the proper data and information that a Level II radiographer is trained to understand. 

PHASE I: Phase I will consist of development of prototype algorithms on representative hardware (to be defined prior to contract award)demonstrated in laboratory simulated environments. The government may also provide actual images obtained during prior government testing. A final report will document testing results and present the top level plan to continue development in Phase II. 

PHASE II: Phase II will continue the success of Phase I and integrate the hardware/software/firmware solution into a representative radiography system at a government facility (to be defined prior to Phase II contract award). The result of Phase II will be a prototype design, including applicable technical data, which will be integrated into current and future radiography inspection systems at multiple government locations. 

PHASE III: Upon success of Phase II, these technologies would be qualified and transitioned to inspection equipment at multiple government ammunition production and R&D facilities. Commercial applications could include medical imaging and inspection of high value and/or safety critical items. 


1: "Automated Defect Recognition and Identification in Digital Radiography", P. Baniukiewicz, Journal of Nondestructive Evaluation, September 2014, Volume 33, Issue 3, pp 327–334

2:  " Automatic Detection of Welding Defects using Deep Neural Network," Wenhui Hou et al, 2018 J. Phys.: Conf. Ser. 933 012006.

3:  "Automatic Defect Recognition in X-Ray Testing Using Computer Vision," D. Mery and C. Arteta, 2017 IEEE Winter Conference on Applications of Computer Vision (WACV), Santa Rosa, CA, 2017, pp. 1026-1035.

4:  "Multiclass classification of weld defects in radiographic images based on support vector machines", Mekhalfa Faiza, and Nafaa Nacereddine. Signal-Image Technology and Internet-Based Systems (SITIS), 2014 Tenth International Conference on. IEEE, 2014.

5:  "Intelligent Segmentation Of Industrial Radiographic Images Using Neural Networks", Lawson, Shaun & Parker, Graham, Proceedings of SPIE - The International Society for Optical Engineering. 10.1117/12.188736, 1994.

6:  - NAS410 NAS Certification & Qualification of Nondestructive Test Personnel

KEYWORDS: Non-destructive Test, Radiography, X-ray, N-ray, Munitions, Testing, Machine Learning, Artificial Intelligence, Inspection 

US Flag An Official Website of the United States Government