Abstract
In a typical painting plant, sealing and cavity waxing operations take place to ensure water tightness and sheet metal edge corrosion protection. Paint defects appear as an inevitable variation of the optimum conditions because of powder, worker mistakes, temperature variation, etc. and show up as surface alterations in the layer they are present. Defects may consist of small particles on the surface, which are difficult to detect. In the case of small particles, their size increase incrementally with each additional paint application until they do not meet the quality control standards, resulting in the rejection of the vehicle. This paper presents the design and construction of a completely automated inspection system (inspection tunnel) for defect detection and monitoring of painted bodies based on computer vision techniques and distributed computation. It is an absolutely new inspection system that has been installed in the Almussafes plant of the FORD Company. The system is able to detect 100% of millimetric defects on large volumes such as car bodies within a time processing less than 10sec.
Keywords: Quality Control, Artificial Vision, Image Processing, Process Automation, Design Automation