| Summary: | In the last decade, there has been considerable growth in the production of end-use polymer parts and components using additive manufacturing methods. A wide range of polymers, from Nylon-12 to thermoplastic polyurethane polymers, can be processed with complex geometry tailored to specific function. However, due to the nature of the layer-by-layer process used in additive manufacturing, high roughness surfaces remain on the parts. To reduce the roughness of the surfaces, a proprietary post-processing method, developed by Additive Manufacturing Technologies, is applied to the surfaces. To monitor and control the finishing of the surfaces, an in-process surface detection instrument has been developed based on machine vision and machine learning. This paper presents the machine learning approach and the effectiveness of the instrument for in-process measurement of the finished surfaces.
|
|---|