Abstract
With the advancement of miniaturization in electronics and the ubiquity of micro-electro-mechanical systems (MEMS) in different applications including computing, sensing and medical apparatus, the importance of increasing production yields and ensuring the quality standard of products has become an important focus in manufacturing. Hence, the need for high-accuracy and automatic defect detection in the early phases of MEMS production has been recognized. This not only eliminates human interaction in the defect detection process, but also saves raw material and labor required. This research developed an automated defects recognition (ADR) system using a unique plenoptic camera capable of detecting surface defects of MEMS wafers using a machine-learning approach. The developed algorithm could be applied at any stage of the production process detecting defects at both entire MEMS wafer and single component scale. The developed system showed an F1 score of 0.81 U on average for true positive defect detection, with a processing time of 18 s for each image based on 6 validation sample images including 371 labels.
Highlights
Modern electronic devices, such as smart phones, consumer electronics, healthcare devices, or surveillance and safety assistant systems, combine a huge variety of functions and offer a high level of comfort and functionality in a reduced space
Micro- and nano-manufacturing and metrology has been driven by micro-electromechanical systems (MEMS), where well-established manufacturing methods based on semiconductor technologies are able to produce structures in micro-meter dimensions [1]
Due to the nature of the MEMS defects, a convolutional neural network (CNN) model could not be used as the core machine learning (ML) algorithm since there were several defects per input image that needed to be localized in addition to being detected and classified
Summary
Modern electronic devices, such as smart phones, consumer electronics, healthcare devices, or surveillance and safety assistant systems, combine a huge variety of functions and offer a high level of comfort and functionality in a reduced space. Micro- and nano-manufacturing and metrology has been driven by micro-electromechanical systems (MEMS), where well-established manufacturing methods based on semiconductor technologies are able to produce structures in micro-meter dimensions [1]. In [6], a new technique based on transient infrared thermography in a transmission mode was used to detect a multilayered MEMS for defect detection. It was concluded that using the aforementioned technique, the size of defects could be estimated more consistently using the surface temperature gradient for transmission mode thermography compared to the reflection mode. This technique would only be able to detect defects such as delamination and voids, with
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
