| URN |
etd-0817121-223136 |
Statistics |
This thesis had been viewed 106 times. Download 3 times.
|
| Author |
Mei-Lan HSU |
| Author's Email Address |
No Public. |
| Department |
Institute of Industrial Management |
| Year |
2020 |
Semester |
2 |
| Degree |
Master |
Type of Document |
Master's Thesis |
| Language |
zh-TW.Big5 Chinese |
Page Count |
53 |
| Title |
Research on Application of Neural Network in the Prediction of Display Optical Quality - The case of OO Company |
| Keyword |
Neural Networks
Optical Quality
TFT-LCD
TFT-LCD
Optical Quality
Neural Networks
|
| Abstract |
Since the end of 2019, the impact of COVID-19 has affcted the economy and changed people's habits and ways of life. The changes in the rise of home office and remote teaching and the increase in medical needs have highlighted the significance of the impact. When significant impacts are on display in the uses of medical treatment, the requirements in quality and safety are higher than in other applications. The optical quality of the display is very important for medical personnel to make high-efficiency and accurate judgments when diagnosing detected images in patients. Therefore, the goal of this research is to use the medical display as the research object. The application of Neural Networks will be used to predict the optical quality of the display in order to reduce the product defect ratio. This application can reduce manhours and damage to the maunfacturing material due to the optical measurement and heavy work. It also consumes less time. So, in turns, product yield should improve and costs be reduced. First, the variables are input into MLP to generate the prediction model which is then compared against the actual measurement results in order to provide analysis and future research directions. The actual verification results show that the Neural Network prediction model constructed by MLP can indeed be used as an effective optical quality prediction method. |
| Advisor Committee |
Chiu-Tang Lin - advisor
Bing-Hou Sun - co-chair
Chiu-Tang Lin - co-chair
Dong-Liang Chen - co-chair
|
| Files |
indicate access worldwide |
| Date of Defense |
2021-06-28 |
Date of Submission |
2021-08-17 |
