Smart measurement system based on deep learning for microstructured surfaces

Yongqiang Yang, Chi Fai Cheung, Da Li

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

Abstract

The microstructured surfaces are playing an important role in industry. Measuring these surfaces quickly and accurately can provide significant quantitative results for machining process. However, traditional measurement methods and systems cannot meet the requirements of high precision machining. Measured samples are often moved several times in order to achieve the desired result. To address these limitations, a smart measurement system based on deep learning algorithm is established. The whole measurement system is portable and can be employed in many works situation. This smart measurement can offer an effective way of capturing high resolution images. To enhance the accuracy and efficiency of data processing, a deep learning model is applied in this system. This model can be employed to generate disparity after the input of high-resolution images. After the information fusion of disparity and the depth, which is obtained from the depth estimation module, the microstructured surfaces can be reconstructed accurately. This system is inputted with the synthetic data and experimental results show that the measurement uncertainty could reach at a sub-micrometer level.

Original languageEnglish
Title of host publicationEighth Asia Pacific Conference on Optics Manufacture and Third International Forum of Young Scientists on Advanced Optical Manufacturing, APCOM and YSAOM 2023
EditorsXuejun Zhang, Xiaoyong Wang, Yifan Dai, Lingbao Kong, Dawei Zhang, Feng Gong, Lihua Li
PublisherSPIE
Volume12976
ISBN (Electronic)9781510672673
DOIs
Publication statusPublished - Aug 2023
Event8th Asia Pacific Conference on Optics Manufacture, APCOM 2023 and 3rd International Forum of Young Scientists on Advanced Optical Manufacturing, YSAOM 2023 - Shenzhen, China
Duration: 4 Aug 20236 Aug 2023

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume12976
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

Conference8th Asia Pacific Conference on Optics Manufacture, APCOM 2023 and 3rd International Forum of Young Scientists on Advanced Optical Manufacturing, YSAOM 2023
Country/TerritoryChina
CityShenzhen
Period4/08/236/08/23

Keywords

  • deep learning
  • information fusion
  • micro-structured surfaces.
  • smart measurement system

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Smart measurement system based on deep learning for microstructured surfaces'. Together they form a unique fingerprint.

Cite this