A novel imaging-enhancement-based inspection method for transparent aesthetic defects in a polymeric polarizer

Abstract The transparent aesthetic defects of polymeric polarizers are difficult to image and characterize using conventional illumination. To inspect such special defects, an automated inspection method was studied that employs the structured lighting technique to markedly enhance defect imaging. The defect was modelled as a microscale plano-convex lens, which has a slight difference in refractive index within its normal region. The imaging enhancement mechanism of the defect was simulated with a model in which the binary stripe patterns of the structured-light source are equivalent to a one-dimensional diaphragm. The experimental results were in agreement with the simulation, suggesting that the model and mechanism are feasible. A novel spatial texture filtering algorithm with higher speed and higher accuracy is proposed to process the structured light images. Approximately 200 samples with defects were successfully imaged, processed and characterized. This inspection method was verified by the final experimental results and has potential for real-time and in situ testing of defects in other polymer films and products.