Tech Briefs

Measuring and Estimating Normalized Contrast in Infrared Flash Thermography

Combining temperature contrast analysis with pixel intensity contrast analysis yields better results in characterizing void-like anomalies.

Infrared flash thermography (IRFT) is used to detect void-like flaws in a test object. The IRFT technique involves heating up the part surface using a flash of flash lamps. The post-flash evolution of the part surface temperature is sensed by an IR camera in terms of pixel intensity of image pixels. The IR technique involves recording of the IR video image data and analysis of the data using the normalized pixel intensity and temperature contrast analysis method for characterization of void-like flaws for depth and width.

This work introduces a new definition of the normalized IR pixel intensity contrast and normalized surface temperature contrast. A procedure is provided to compute the pixel intensity contrast from the camera pixel intensity evolution data. The pixel intensity contrast and the corresponding surface temperature contrast differ but are related. This work provides a method to estimate the temperature evolution and the normalized temperature contrast from the measured pixel intensity evolution data and some additional measurements during data acquisition.

Thermal simulation software, such as Thermo-Calc, provides simulation of surface temperature evolution on void-like flaws. A comparison of the experimentally estimated temperature contrast and simulation estimated temperature contrast is required to validate the simulation model and its input parameters. Conversely, if the simulation temperature contrast data is available, then a method provided here can be used to estimate the pixel intensity and the pixel intensity contrast based on known values of parameters of the data acquisition setup.

The normalized pixel intensity contrast and the normalized temperature contrast differ for objects with emissivity other than 1. Therefore, for better contrast analysis the two quantities should not be treated as the same. To compare the simulation temperature contrast with the measured pixel contrast, it is necessary to estimate the reflection temperature evolution. It is also necessary to estimate the incident heat flux. Ideally, the simulation should model the compound heat source flux evolution, which also includes the post-flash thermal afterglow. The effect of reflection temperature in the pixel intensity also should be accounted for to seek a better estimation of the temperature contrast evolution from the pixel intensity evolution data.

Using formulas given here, the reflection temperature evolution and the temperature contrast evolution can be estimated from the IRFT data. An emissivity factor, defined here, relates the temperature contrast to the pixel intensity contrast.

Reflection temperature evolution can be used to model the afterglow flux of the flash source in the simulation to estimate the temperature contrast evolutions and the pixel intensity contrast evolution on simulated voids.

An emissivity estimation technique was developed using the IR camera. If the IR camera is programmed with the reflection temperature formulas derived here, the camera can provide the object surface temperature directly even during the IRFT data acquisition. The IR camera can be programmed to estimate the object emissivity in real time using the technique provided here. Due to improvement in the contrast analysis, including the modeling of source in thermal simulation, void-like anomalies are characterized more precisely.

This work was done by Ajay M. Koshti of Johnson Space Center.

This invention is owned by NASA, and a patent application has been filed. Inquiries concerning nonexclusive or exclusive license for its commercial development should be addressed to the Patent Counsel, Johnson Space Center, (281) 483-1003. MSC-24506-1

This Brief includes a Technical Support Package (TSP).

Measuring and Estimating Normalized Contrast in Infrared Flash Thermography (reference MSC-24506-1) is currently available for download from the TSP library.

Please Login at the top of the page to download.