An automatic procedure to code the inspection guideline for vehicle headlamp lenses

Nowadays the quality control of headlamp lenses for vehicles is visually made by expert operators, known as inspectors. To establish a standard in this manual process, the inspectors must follow the inspection guideline. This guideline is a document, defined by experts in the inspection of lenses and validated by the customer, that presents the criterions of acceptance and rejection for the flaws that may appear on the lens surface. This paper proposes a methodology to include the inspection guideline in an automated headlamp lens inspection system based on computer vision. As the way in which the guideline includes the knowledge of an expert in the inspection of lenses is inherently qualitative and vague, a fuzzy rule-based system is developed to model this information.