Pattern recognition for MEMS images of surface topography using wavelets

Hu Kai, Jiang Xiangqian, Liu Xiaojun

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


Micro-Electro-Mechanical Systems (MEMS) have been growing in interest in recent years. Due to the small size of the MEMS, the traditional method of metrology measurement seriously affects the parameter of the object being measured, and high accuracy metrology cannot be acquired. MEMS microstructure images of surface are most composed of a number of steps, grooves and slots. Pattern analysis and recognition of these microstructures with linear feature is one of the key problems in metrology and testing technology of Micro-Electro-Mechanical-System (MEMS). Effective detections of these components play an important role in simplifying feature model in pattern recognition and pattern match. Traditional Linear feature detectors based on pixel processing each by each may fail to detect out lines in image with low SNR. A fast discrete beamlet transform and a novel method of linear feature detection are proposed, which can detect lines with any orientation, location and length. Experiment results prove the efficiency of the method proposed even in image with very low SNR.

Original languageEnglish
Title of host publicationMIPPR 2007
Subtitle of host publicationRemote Sensing and GIS Data Processing and Applications; and Innovative Multispectral Technology and Applications
EditorsYongji Wang, Bangjun Lei, Jing-Yu Yang, Jun Li, Chao Wang, Liang-Pei Zhang
Number of pages5
ISBN (Print)9780819469540
Publication statusPublished - 14 Nov 2007
EventInternational Symposium on Multispectral Image Processing and Pattern Rrecognition - Wuhan, China
Duration: 15 Nov 200717 Nov 2007

Publication series

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


ConferenceInternational Symposium on Multispectral Image Processing and Pattern Rrecognition


Dive into the research topics of 'Pattern recognition for MEMS images of surface topography using wavelets'. Together they form a unique fingerprint.

Cite this