Neural Learning Based Blind Source Separation for Detection of Material Defects in Pulsed Thermography Images
In this paper, we introduce a neural learning-based approach to blind source separation for detection of material flaws in pulsed thermography (PT) images. This approach can be used to detect internal defects (pores) in metallic Additively Manufactured (AM) materials. Such defects occur in high-strength alloys produced with direct laser sintering AM method for nuclear energy applications. Pulsed thermal imaging system utilizes a high intensity flash lamp to rapidly heat surface of sample, a high sensitivity infrared camera to capture data of surface temperature variations. The data cube obtained with PT (stack of surface temperature images at different times) can be analyzed with image processing algorithms to detect material defects. Compared with conventional nondestructive evaluation (NDE) methods, such as digital radiography and ultrasonic testing, the PT pulsed infrared thermal detection technique has advantages of one-sided non-contact measurements, fast processing of large sample areas captured in one image. Detection of small material defects requires finding features in thermal images, which have signal contrast levels approaching sensitivity limit of IR camera. In this study, an optimized Neural Learning based Blind Source Separation (NLBSS) algorithm, including Principal Component Analysis (PCA), and Independent Component Analysis (ICA) is demonstrated to automatically extract principal temporal and spatial features of thermography frames to enhance flaw detection. By using the NLBSS algorithm, material internal defects can be automatically detected. Furthermore, this processing approach compensates for experimental thermal imaging artifacts, such as noise and uneven heating. By merging artificial intelligence with the photothermic, the NDE system detects internal calibrated defects of various sizes and depths in AM nuclear-grade metallic alloys.
- Research Organization:
- Argonne National Lab. (ANL), Argonne, IL (United States)
- Sponsoring Organization:
- USDOE Office of Nuclear Energy - Nuclear Energy Enabling Technologies (NEET)
- DOE Contract Number:
- AC02-06CH11357
- OSTI ID:
- 1804074
- Resource Relation:
- Conference: 20th Annual IEEE International Conference on Electro Information Technology, 05/28/20 - 05/30/20, Naperville, IL, US
- Country of Publication:
- United States
- Language:
- English
Similar Records
Performance of Compact Pulsed Thermal Imaging System for In-Service Applications. Pulsed thermal tomography nondestructive examination of additively manufactured reactor materials and components
Pulsed Thermal Tomography Nondestructive Examination of Additively Manufactured Reactor Materials and Components (Final Technical Report)