The robust detection and removal of unwanted asperities in applied surface metrology

The presence of unwanted asperities or debris in surface metrology data is a problem faced by many manufacturers of high volume precision components. The process of re-cleaning and re-measurement upon detection of an asperity can be extremely time consuming and costly. Current asperity removal methodologies are significantly limited due their fixed formats and to the user selection of thresholds. This paper presents an adaptive method of detection and an accompanying procedure for padding across the removal area. Results are presented for asperity removal in the measurement of roundness on fuel injector components.