Image-Based Parametric Pattern Recognition for Micro- and Nano- Defect Detection. Issue 2 (2020)

• Record Type: Journal Article
• Authors: Belikov, Sergey; Su, Chanmin; Enachescu, Marian
• Is Part Of: IFAC-PapersOnLine. Volume 53:Issue 2(2020)
• Page Start: 8591
• Page End: 8598
• Publication Date: 2020
• Subjects: Micro- -- Nano-Mechatronics Systems -- Defect Detection -- Pattern Recognition -- Modeling; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2020.12.528
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: Detection and characterization of defects at micro- and nano-scale is one of the most important tasks of the production process control technologies at these scales. Examples include flat panel displays (FPD) and MEMS (defects at micro scale), and hard disk drive substrates up to 95 mm in diameter (defects at nano scale). Although different measurement technologies are applied, such as infrared thermography at micro-scale and Atomic Force Microscopy (AFM) at nano-scale, a similar technique of image-based defect recognition and location can be utilized, namely parametric pattern recognition, where the defect is defined by a specific set of inequalities in the space of selected measurable variables. We describe the technique and demonstrate it for the following applications: 1) mechanical defects (high friction) in MEMS; 2) electrical defect locations (shorts, opens) on FPD; and 3) nano-asperities on hard disk drive surfaces. The first two use infrared thermography, and the third uses multi-channel AFM scanning. Both large and small fields of view images are needed for analysis.