Robotic Paint Defects Detection Uses Stripe Reflectometry

Autaza Paint is an optical quality inspection system for capable of identifying over 30 types of appearance defects in plastic, metal, and various composite materials. The optical sensing system uses stripe reflectometry technology to highlight defects. The Autaza Paint measurement head has a monitor with high brightness, frequency and resolution, and a high-resolution industrial camera. The rugged composition means that the system does not need external light and can work in environments with ambient lighting up to 700 lumens. The components have a specific position and angulation to acquire the images. For each inspected area, 4 pictures are taken by translating the white stripes, thus ensuring that the area is completely inspected.

The patented Zebra Analysis method uses a set of lights with geometric patterns transmitted over a part. A dedicated system for automatic inspection captures images of the part with the projected lighting. The defects detection is performed automatically by analyzing the distortions caused in zebra pattern.

The system has intelligent software that uses computer vision and artificial intelligence to identify and classify paint defects. The system is calibrated according to the client’s quality criteria and can have a binary response (defect and non-defect) or up to 3 levels of severity of defects (low, medium, and severe).

The number of robots installed with Autaza Paint heads in an inspection cell is determined by the cycle time of the production line. The system works with any brand of robot that supports the payload, degrees of freedom, and reach requirements.

Autaza emerged from a 3-year research project between university (Aeronautics Institute of Technology, ITA) and the automotive industry (General Motors Brazil). Recognizing the innovation potential of the developed prototypes Autaza was founded in 2016, with GM as its first customer.

For more information: