System for detecting adhesive overflow in sensor chips based on improved Unet

Lei Chen; Zourong Long; Bin Tang; Mingfu Zhao; Yue Leisi; Yulong He; Fuping Wei

doi:10.1117/12.3025172

8 February 2024 System for detecting adhesive overflow in sensor chips based on improved Unet

Lei Chen, Zourong Long, Bin Tang, Mingfu Zhao, Yue Leisi, Yulong He, Fuping Wei

Proceedings Volume 13066, International Conference on Optoelectronic Materials and Devices (ICOMD 2023); 1306615 (2024) https://doi.org/10.1117/12.3025172
Event: 2023 International Conference on Optoelectronic Materials and Devices (COMD 2023), 2023, Chongqing, China

Abstract

With the rapid development of information technology, traditional neural networks used as feature extraction networks can improve the network’s fitting ability but may lose information for small object detection, resulting in low accuracy. In this paper, the image acquisition device and Unet detection model were built independently. The algorithm accurately detects the sensor chip overflow using image processing techniques with OpenCV. Finally, the detected images are presented using PyQt.Experimental results show that the improved Unet-glue algorithm achieves better segmentation accuracy for chip overflow. It also demonstrates strong robustness and practicality in the field of small object defect detection.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Lei Chen, Zourong Long, Bin Tang, Mingfu Zhao, Yue Leisi, Yulong He, and Fuping Wei "System for detecting adhesive overflow in sensor chips based on improved Unet", Proc. SPIE 13066, International Conference on Optoelectronic Materials and Devices (ICOMD 2023), 1306615 (8 February 2024); https://doi.org/10.1117/12.3025172

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available