返回
Multi-Label Separation-Deviation Surface Model for Detecting Spatial Defects in Topographic Surfaces
DOI:10.1109/TII.2020.3020051.png)
摘要
En 中文
Surface topography is a critical quality characteristic of many products and manufacturing processes. Various defects commonly appear on the topography of finished products in spatial patterns after or during manufacturing. These defects are difficult to be identified using traditional monitoring approaches because of the complex structure of topographic data. This article develops a novel and effective approach for monitoring spatial defects in topographic surfaces. The approach improves the representation of surface characteristics through the developed multilabel separation-deviation surface (MSS) model, which labels the important surface characteristics and smooths out the noisy characteristics. We develop two features for monitoring changes in the characteristics of the assigned labels. The MSS feature is introduced for capturing deviations within the assigned labels, and the generalized spatial randomness feature is derived for quantifying deviations between the assigned labels. These two features are integrated into a single monitoring statistic, which is successfully applied for detecting various defects in topographic surfaces, outperforming the traditional monitoring approaches.
Keyword:
Manufacturing processes
Feature extraction
Surface topography
Noise measurement
Surface finishing
Monitoring
Surface treatment
Anomaly detection
separation-deviation (SD) model
spatial randomness (SR)
surface monitoring
surface topography
AI总结
对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
9.9
论文数:
8.3K
被引数:
6.0W
机构
引用论文
Effects of Grinding Passes and Direction on Material Removal Behaviours in the Rail Grinding Process钢轨打磨过程中打磨道次和方向对材料去除行为的影响
MATERIALS
IF3.2

