Article ID Journal Published Year Pages File Type
529858 Pattern Recognition 2015 9 Pages PDF
Abstract

•We present a thorough analysis of reflection properties of fibrous materials.•We present a photometric stereo method for inspection of carbon fiber surfaces.•We describe an algorithm for calculation of the direction of carbon fibers.•We present results of six experiments that demonstrate the accuracy of the system.

Carbon- and glass fibre materials exhibit challenging optical properties, in particular highly specular reflectivity. State of the art vision-based sensor systems use diffuse light setups to suppress specular reflections and apply texture analysis to obtain fibre orientations for surface patches.We investigate the reflection behaviour of fibrous materials. The derived reflection model is the basis for a sensor system that directly measures fibre orientation as well as diffuse and specular reflectivity per pixel. The proposed sensor is robust to changes of the fibre material and capable to deliver fast and accurate information about fibre direction without time consuming texture analysis. An empirical evaluation shows that the root mean squared error of calculated fibre angles is around or below 1°.

Related Topics
Physical Sciences and Engineering Computer Science Computer Vision and Pattern Recognition
Authors
, , , ,