A real-time defective pixel detection system for LCDs using deep learning based object detectors
- PDF / 1,530,554 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 110 Downloads / 321 Views
A real-time defective pixel detection system for LCDs using deep learning based object detectors Aslı Çelik1 · Ayhan Küçükmanisa1
· Aydın Sümer1 · Aysun Ta¸syapı Çelebi1 · Oguzhan ˘ Urhan1
Received: 9 January 2020 / Accepted: 21 October 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract The presence of pixel defects on the screens of LCD-based products (TV, tablet, phone, etc.) is unacceptable given the consumer expectations. Therefore, these defects should be detected before the product reaches the user during the production stage. Visual inspections are mostly performed by human operators in the production. These inspections are error prone and not efficient in terms of consumed time. For this reason, computer visionbased approaches are started to find applications in this kind of problems. This paper presents an image acquisition system and a detailed analysis of deep learningbased object detectors for LCD pixel defect detection problem. Experimental results show that the proposed methods can be a powerful alternative to operator control by providing more efficient use of time, human, financial resources and betterquality standards in TV production industry. Keywords Pixel defect · Defect detection · LCD · Deep learning
Introduction Over the past decade Liquid Crystal Displays (LCDs) using Thin Film Transistor (TFT) arrays have become very popular display devices used in notebooks, personal computers, mobile phones, monitors and televisions. Because TFT-LCD panels have many advantages over the traditional CRT monitors such as fullcolor display capabilities, smaller space, lower cost, radiation, and power consumption (Li et al. 2012, Chen et al. 2013). With the increase in demand of TFTLCD panels, the control of visual quality has been a critical and important task for LCD manufacturers to guarantee high quality control standards. Because of the complex manufacturing process of TFT-LCD panels, small particles of dirt and manufacturing process itself, originating from production equipment inconsistencies or from human mistakes, some LCDs may have several defects such as line, point and Mura defects. The inspection process in many LCD panel production is performed visually by human test operators. However, human vision-based inspection processes have some disadvantages such as the higher labor cost, lower inspection
B 1
Ayhan Küçükmanisa [email protected] Faculty of Engineering, Electronics and Communication Engineering, Kocaeli University, Kocaeli, Turkey
efficiency, and subjective inspection results based on operator. Additionally, it is difficult to find defects with naked eye by human test operators. Hence, automated inspection systems for LCD panels are necessary in order to increase productivity and reliability, reducing the equipment costs of the devices (Juen et al. 2019). Today many LCD panel manufacturers have started to use automated defect inspection systems by using image processing, and machine learning based approaches (Kim et al. 2020). The LCD defect
Data Loading...
