A study on spline filter that achieve both robustness and lower compatibility

2022 Fiscal Year Final Research Report

• Project/Area Number: 20K04202
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 18020:Manufacturing and production engineering-related
• Research Institution: Chukyo University
• Principal Investigator: Numada Munetoshi 中京大学, 工学部, 教授 (00554924)
• Project Period (FY): 2020-04-01 – 2023-03-31
• Keywords: ロバスト性 / 上位互換性 / スプラインフィルタ / 表面性状

Outline of Final Research Achievements

A spline filter is an epoch-making low-pass filter for roughness that can measure both ends of a measurement object. Furthermore, a robust spline filter, which is not affected by outliers, was drafted by ISO as a universal filter, and was installed one after another in surface roughness measuring machines around the world. However, when applied to measurement data with no anomalies, it was found that the output was completely different from that of a normal spline filter, and the lack of lower compatibility of the filter became apparent.
In this research, we established a new algorithm to apply the compatibility of robustness and lower compatibility, which was first achieved by the principal investigators in the Gaussian filter, a typical low-pass filter, to spline filters, and solved the problem in the field of measurement.

Free Research Field

機械計測

Academic Significance and Societal Importance of the Research Achievements

本研究は研究代表者らが開発したロバスト推定法である高速M推定法を、同じく研究代表者らが考案した畳み込み演算方式のスプラインフィルタに適用する手法である。これによりロバスト性と上位互換性を備えたスプラインフィルタを実現でき、さらにこのフィルタ演算を周波数領域で行うことで高速化した。本研究の遂行により、学術的には逆行列型を含むあらゆるロバストフィルタで上位互換性の保証を可能にできる手法が提案できた。