Contrast based imaging and analysis computer-implemented method to analyze pulse thermography data for nondestructive evaluation
Inventors
Assignees
National Aeronautics and Space Administration NASA
Publication Number
US-10242439-B1
Publication Date
2019-03-26
Expiration Date
2037-08-16
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
Methods and systems for analyzing and processing digital data comprising a plurality of infra-red (IR) video images acquired by a pulse thermography system are used to compute video data from the raw and smoothed video data acquired for the performance of non-destructive evaluation. New video data types computed may include but are not limited to contrast evolution data such as normalized contrast, converted contrast and normalized temperature contrast. Additionally, video data types computed comprise surface temperature, surface temperature rise and temperature simple contrast.
Core Innovation
The invention provides methods and systems for analyzing and processing digital data comprising a plurality of infra-red (IR) video images acquired by a pulse thermography system to compute new video data from the raw and smoothed video data. This computation includes contrast evolution data such as normalized contrast, converted contrast, and normalized temperature contrast, as well as other video data types like surface temperature, surface temperature rise, and temperature simple contrast. The methods detail steps for creating and analyzing these video data types to enhance non-destructive evaluation.
The invention addresses the problem of detecting and characterizing subsurface anomalies, such as delaminations in thin, nonmetallic materials used in aerospace, via infrared flash thermography. Traditional IR flash thermography captures surface temperature changes after a heat pulse. However, anomalies require advanced video data processing methods to increase detection sensitivity and accurately characterize anomaly properties including size and depth. Existing methods lacked systematic contrast-based video processing, derivative computation, and calibration to improve anomaly detection and depth evaluation.
The invention introduces three contrast video image processing (CVIP) methods: the Normalized Contrast and Derivatives (NCD) method, the Converted Contrast and Derivatives (CCD) method, and the Normalized Temperature Contrast and Derivatives (TCD) method. Each method operates independently on the IR video data to compute contrast evolutions and their derivatives, extract feature images (A-scan, B-scan, C-scan analogs), and perform depth and size evaluations. Further, an empirical Contrast Evolution Calibration and Analysis (CECA) method employs calibration standards with known artificial flaws to provide contrast evolution parameters that enable depth and anomaly characterization. Additional image and data analysis techniques such as edge detection, image profiling, image registration, mosaic creation, and probability of detection (POD) analysis are integrated.
Claims Coverage
The claims include one independent claim encompassing multiple inventive features related to video data creation and analysis for infrared flash thermography.
Creation of normalized temperature contrast video data and derivatives
The method selects frames of infrared flash thermography video comprising surface temperature pixel intensity sequences after heat flash. It computes normalized temperature contrast video data and corresponding first and second time derivatives using specified equations, optionally utilizing raw or smoothed video data.
Creation of extracted images and frame images from contrast data
The method combines image data from multiple frames or selected frames associated with peak, minimum, averages, threshold values, or pixel lines for normalized temperature contrast and its derivatives to produce extracted images and frame images representing anomaly features.
Use of calibration standard for calibration data generation and anomaly characterization
The method employs a calibration standard with defined flat bottom holes or embedded gaps with known diameters and depths. It generates calibration data including known diameters, depths, and multiple normalized contrast evolution parameters. This calibration data is used to characterize anomaly depth and diameter in test materials.
Computerized flaw size measurement and monitoring methods
The method includes measuring flaw size on video frames using superimposed lines with computerized distance measurement, automated edge detection for flaw sizing, and monitoring flaw growth by comparing reference images with subsequent images via registration, superimposition, or subtraction.
Use of contrast data and diameter-to-depth correlations for probability of detection (POD) analysis
The method utilizes normalized temperature contrast evolution parameters and diameter-to-depth ratios as inputs to calculate probability of detection of flaws. It determines detection thresholds and correlates thermography response with flaw size and depth for POD evaluations.
The independent claim encompasses a comprehensive method for creating normalized temperature contrast video data with derivatives from IR flash thermography data, producing various image representations, employing calibration standards for anomaly characterization, measuring flaw size with computer assistance, and performing probability of detection analyses based on contrast and flaw size correlations.
Stated Advantages
Enhanced detection and characterization of subsurface anomalies through three independent contrast video image processing methods (NCD, CCD, TCD) enabling extraction of detailed contrast evolution features and derivatives.
Reduction of noise and improved image sharpness through smoothing and derivative computations, allowing better visualization and sizing of anomalies.
Use of an empirical calibration standard and associated contrast parameters permits quantitative assessment of anomaly depth and size.
Converted contrast method provides advantage of eliminating or reducing reliance on a reference region of interest, improving graphical display and detection robustness.
Normalized temperature contrast reduces influence of diffused reflection and yields more quantitative surface temperature measurements compared to pixel intensity, enhancing anomaly contrast and detection.
Integration of image analysis methods including edge detection, image profiling, and image registration facilitates automated flaw sizing and monitoring of material condition changes over time.
Methods enable establishing accept/reject thresholds for thermography response to support quantitative flaw detection and probability of detection (POD) analysis, thereby facilitating more reliable NDE assessments.
Documented Applications
Non-destructive evaluation (NDE) of thin nonmetallic materials such as laminated or bonded composites in aerospace for detection and characterization of delaminations and surface cracks using infrared flash thermography.
Calibration of flash thermography systems using standards with artificial flaws (flat bottom holes or embedded gaps) for precise depth and diameter evaluation of anomalies.
Flaw size measurement and monitoring for damage growth assessment in composite materials by producing, registering, and comparing thermography images over time.
Quantitative flaw detection through thermography response-based accept/reject thresholding and probability of detection analysis correlating thermography response with flaw size and depth.
Interested in licensing this patent?