DocumentCode :
1785314
Title :
Automated HDRI solves challenge of detecting defects in highly reflective fibers
Author :
Wilson, Douglas H.
Author_Institution :
FiberQA, LLC, Old Lyme, CT, USA
fYear :
2014
fDate :
11-13 Nov. 2014
Firstpage :
85
Lastpage :
86
Abstract :
A method for handling wide range of reflectance between the core and cladding due to the on-axis inspection lighting reflecting off the distal end of the fiber using high dynamic range (HDR) imaging has been demonstrated using the FastMT software. A set of fibers was imaged using two different exposure levels to create two low dynamic range (LDR) images. These images were combined to form an uncompressed HDR image. Due to current limitation of the bit-depth of the processing, a compression technique was applied to convert the HDR image to an 8-bit range. The resulting detection of defects in both the cladding and core, from a single image acquisition, proves that the method will provide excellent fiber visual pass/fail results. To the best of our knowledge, this is the first fiberscope processing software to employ HDR imaging and processing to handle extreme dynamic range cases. Further development is planned to enable direct processing of the HDR image without the loss in contrast of non-linear compression used in this result. It is anticipated that this improvement will recover at least a factor of two in contrast, making it possible to reliably detect even less visible defects.
Keywords :
data compression; image coding; optical fibre cladding; optical fibre testing; 8-bit range; FastMT software; LDR images; automated HDRI; bit-depth; cladding; compression technique; core; defect detection; extreme dynamic range cases; fiber visual pass-fail results; fiberscope processing software; highly reflective fibers; low dynamic range images; nonlinear compression; on-axis inspection lighting; reflectance; single image acquisition; Dynamic range; Image coding; Imaging; Inspection; Optical fiber cables; Optical fiber devices; Reflectivity;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Avionics, Fiber-Optics and Photonics Technology Conference (AVFOP), 2014 IEEE
Conference_Location :
Atlanta, GA
Print_ISBN :
978-1-4799-2473-8
Type :
conf
DOI :
10.1109/AVFOP.2014.6999436
Filename :
6999436
Link To Document :
بازگشت