The major advantage of products made from composite materials is given by their superior weight-specific mechanical properties. These can be weakened by defects induced in the manufacturing process. Therefore, online detection and analysis of the processed fiber bundle geometry is a key factor for the quality assurance of the final part. In this article, the instrumentation and data evaluation for determining the surface geometry of fiber bundles by means of light sectioning was examined. Bundles of glass and carbon fibers were measured continuously on an inspection test rig. Different background materials have been used in order to validate the applicability of the approach. By utilization of a polynomial fitting algorithm, data segmentation of object and baseline was robustly achieved. By means of cross correlation, the data alignment could be evaluated faster and more reliable compared to a method previously presented by us. The information could then be used for the determination of the fiber bundle width, centerline, spatial changes, and oscillations. In addition, unwanted defects as well as lateral movement of the fiber bundles were reliably detected. The information revealed by the proposed algorithm provides the basis for robust online monitoring of fiber bundle geometry in highly automated composite manufacturing processes.