Grewal, G. S., Dubey, V. N. and Claremont, D., 2006. Isochromatic demodulation by fringe scanning. Strain, 42 (4), pp. 273-281.
Full text available as:
Restricted to Registered users only
Official URL: http://www.blackwell-synergy.com/doi/abs/10.1111/j...
RGB calibration is the fastest isochromatic demodulation technique, as it does not require unwrapping as in the case of phase-shifting method. The technique is based on constructing a look up table (LUT) of fringe order against RGB triplet from digital images and decoding the test image from the LUT. Research has shown that the technique, though very fast, is limited to fringe orders up to three with the conventional white light. The colours tend to merge beyond that and make it difficult to obtain a unique value of RGB triplet. Changes in fringe gradient caused by stretching/bunching of fringes in test model further add to errors. Special light sources with narrow-band spectral response are required with fringe tracking algorithms to demodulate higher fringe orders. The calibration technique is also sensitive to geometric and chromatic variations. This paper presents a cost-effective alternative solution to conventional RGB technique using a flatbed scanner. The system is capable of demodulating higher fringe order and incorporates information from other colour spaces. It does not require separate light sources and cameras, and is found to be insensitive to geometric and chromatic variations. Curve fitting technique has been proposed to determine accurate fringe orders.
|Additional Information:||This a fundamental engineering paper that forms part of our programme to develop a shear force sensor for medical applications.|
|Uncontrolled Keywords:||fringe demodulation, image unwrapping, LUT-based calibration, RGB photoelasticity, stress analysis|
|Subjects:||Technology > Manufacturing and Design > Manufacturing|
|Group:||Faculty of Science and Technology|
|Deposited By:||INVALID USER|
|Deposited On:||26 Apr 2007|
|Last Modified:||10 Sep 2014 15:39|
Downloads per month over past year
|Repository Staff Only -|