Reliability-Enhanced Camera Lens Module Classification Using Semi-Supervised Regression Method
Artificial intelligence has become the primary issue in the era of Industry 4.0, accelerating the realization of a self-driven smart factory. It is transforming various manufacturing sectors including the assembly line for a camera lens module. The recent development of bezel-less smartphones necess...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-05-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/10/11/3832 |
| _version_ | 1827715887731310592 |
|---|---|
| author | Sung Wook Kim Young Gon Lee Bayu Adhi Tama Seungchul Lee |
| author_facet | Sung Wook Kim Young Gon Lee Bayu Adhi Tama Seungchul Lee |
| author_sort | Sung Wook Kim |
| collection | DOAJ |
| description | Artificial intelligence has become the primary issue in the era of Industry 4.0, accelerating the realization of a self-driven smart factory. It is transforming various manufacturing sectors including the assembly line for a camera lens module. The recent development of bezel-less smartphones necessitates a large-scale production of the camera lens module. However, assembling the necessary parts of a module needs much room to be improved since the procedure followed by its inspection is costly and time-consuming. Consequently, the collection of labeled data is often limited. In this study, a reliable means to predict the state of an unseen camera lens module using simple semi-supervised regression is proposed. Here, an experimental study to investigate the effect of different numbers of training samples is demonstrated. The increased amount of data using simple pseudo-labeling means is shown to improve the general performance of deep neural network for the prediction of Modulation Transfer Function (MTF) by as much as 18%, 15% and 25% in terms of RMSE, MAE and R squared. The cross-validation technique is used to ensure a generalized predictive performance. Furthermore, binary classification is conducted based on a threshold value for MTF to finally demonstrate the better prediction outcome in a real-world scenario. As a result, the overall accuracy, recall, specificity and f1-score are increased by 11.3%, 9%, 1.6% and 7.6% showing that the classification of camera lens module has been improved through the suggested semi-supervised regression method. |
| first_indexed | 2024-03-10T19:29:08Z |
| format | Article |
| id | doaj.art-da452a83cc1342d0aae903aafb1754eb |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T19:29:08Z |
| publishDate | 2020-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-da452a83cc1342d0aae903aafb1754eb2023-11-20T02:21:37ZengMDPI AGApplied Sciences2076-34172020-05-011011383210.3390/app10113832Reliability-Enhanced Camera Lens Module Classification Using Semi-Supervised Regression MethodSung Wook Kim0Young Gon Lee1Bayu Adhi Tama2Seungchul Lee3Department of Mechanical Engineering, Pohang University of Science and Technology, Pohang 37673, KoreaSamsung Electro-Mechanics, Suwon 16674, KoreaDepartment of Mechanical Engineering, Pohang University of Science and Technology, Pohang 37673, KoreaDepartment of Mechanical Engineering, Pohang University of Science and Technology, Pohang 37673, KoreaArtificial intelligence has become the primary issue in the era of Industry 4.0, accelerating the realization of a self-driven smart factory. It is transforming various manufacturing sectors including the assembly line for a camera lens module. The recent development of bezel-less smartphones necessitates a large-scale production of the camera lens module. However, assembling the necessary parts of a module needs much room to be improved since the procedure followed by its inspection is costly and time-consuming. Consequently, the collection of labeled data is often limited. In this study, a reliable means to predict the state of an unseen camera lens module using simple semi-supervised regression is proposed. Here, an experimental study to investigate the effect of different numbers of training samples is demonstrated. The increased amount of data using simple pseudo-labeling means is shown to improve the general performance of deep neural network for the prediction of Modulation Transfer Function (MTF) by as much as 18%, 15% and 25% in terms of RMSE, MAE and R squared. The cross-validation technique is used to ensure a generalized predictive performance. Furthermore, binary classification is conducted based on a threshold value for MTF to finally demonstrate the better prediction outcome in a real-world scenario. As a result, the overall accuracy, recall, specificity and f1-score are increased by 11.3%, 9%, 1.6% and 7.6% showing that the classification of camera lens module has been improved through the suggested semi-supervised regression method.https://www.mdpi.com/2076-3417/10/11/3832semi-supervised regressioncamera lens modulepseudo-labeldeep neural networkmodular transfer function |
| spellingShingle | Sung Wook Kim Young Gon Lee Bayu Adhi Tama Seungchul Lee Reliability-Enhanced Camera Lens Module Classification Using Semi-Supervised Regression Method Applied Sciences semi-supervised regression camera lens module pseudo-label deep neural network modular transfer function |
| title | Reliability-Enhanced Camera Lens Module Classification Using Semi-Supervised Regression Method |
| title_full | Reliability-Enhanced Camera Lens Module Classification Using Semi-Supervised Regression Method |
| title_fullStr | Reliability-Enhanced Camera Lens Module Classification Using Semi-Supervised Regression Method |
| title_full_unstemmed | Reliability-Enhanced Camera Lens Module Classification Using Semi-Supervised Regression Method |
| title_short | Reliability-Enhanced Camera Lens Module Classification Using Semi-Supervised Regression Method |
| title_sort | reliability enhanced camera lens module classification using semi supervised regression method |
| topic | semi-supervised regression camera lens module pseudo-label deep neural network modular transfer function |
| url | https://www.mdpi.com/2076-3417/10/11/3832 |
| work_keys_str_mv | AT sungwookkim reliabilityenhancedcameralensmoduleclassificationusingsemisupervisedregressionmethod AT younggonlee reliabilityenhancedcameralensmoduleclassificationusingsemisupervisedregressionmethod AT bayuadhitama reliabilityenhancedcameralensmoduleclassificationusingsemisupervisedregressionmethod AT seungchullee reliabilityenhancedcameralensmoduleclassificationusingsemisupervisedregressionmethod |