Measurement Principles of Shearography

General Principle

The very basic idea with Shearography is to look at the test surface of the specimen with a shearography camera. The camera records one interferometric photo of the surface. This photo can be thought of as a unique footprint of this surface, at this unloaded state, including surface roughness and shape.

We now stress this material with a small amount of load, for example with heat. The material wants to expand when heated up, and if it has weak spots it will be allowed to expand more. At the loaded state one more interferometric photo is taken. Now we also have a interferometric footprint of the area at the deformed state.

To extract information about the difference between the two states, we subtract the two images and a shearogram is created.
This sheaogram can be thought of as a topography of the surface, but only gradients (slopes) are measured, not absolute heights of the hills. Your defects will be seen as “hills” popping out of the plane. You can quantify the size of the defects (in plane size) by measuring how large your hills are.

Speckle and Shearing

When a surface is illuminated with a coherent laser light source a stochastic interference pattern is created. This ’speckle’ pattern is projected on a camera’s CCD chip. In contrast to ESPI (Electronic Speckle Pattern Interferometry) where the speckle is compared with a reference light path, Shearography uses a reference created by shearing the image of the test object to create a double image. This makes the method much less sensitive to external vibrations and noise, and you can measure without any need for a vibration damped table, - Shearography is made for field usage! The superposition of the double image, a shearogram, represents the surface of the test object in an unloaded state, (this image can be thought of as an interferometric foot print). By inducing a small strain in the material using thermal, pressure or mechanical loading, the material deforms. If the material has non-homogeneous properties, the deformation of the surface will not be uniform – more deformation at weak areas! A new shearogram is recorded in the now loaded state and is compared with the unloaded image. If a flaw is present it will be seen in this result as a small deformation.


Image 1: Shearography fundamentals; a intereferometric footprint is created from the surface at two states - unloaded test object, and loaded test object. In-structure defects deformations under load will show up when comparing the two foot prints.


Image 2: A primitive Shearography setup; Two physical points on test object will be projected on to one point on the CCD chip to record an interferometic footprint. The tested surface is illuminated with a monochromatic light, typical 650nm.

Image Subtraction

The basic Shearography principle is to subtract two images (interferometric footprints) of a test object; before and after load. Thereafter the information (intensity) from those images are subtracted, and then the surface deformation can be displayed due to the speckle information. Surface roughness will be neglected ad-hoc in the subtraction process.


Image 3: The primitive Shearography principle. A shearography image is recorded at unloaded state and one image is recorded in the loaded state. Thereafter they are subtracted and in the result defects can be detected.

Phase Shift Technology

To increase the sensitivity of the technique, a real-time phase shift process is used in the sensor. This uses a stepping mirror that shifts the reference beam and enhances the results with directional information included with the deformation.


Image 4: A phase stepping Shearography setup; Two physical points on test object will be projected on to one point on the CCD chip to record a interferometic footprint. A modified Michelson cube is here used where with a double breaking mirror as a beam splitter. One mirror is for adjustment of shear properties and the other one is the phase stepper.


Image 5: The phase stepper moves through its 4 positions with a internal difference of 1/4 Wavelength, at each position a image is recorded and sent through the software processor to evaluate the phase relationship with a best fit algorithm.