• IEMEK Journal of Embedded Systems and Applications
• /
• v.17 no.3
• /
• pp.129-138
• /
• 2022

Collecting and labeling sufficient training data, which is essential to deep learning-based visual inspection, is difficult for manufacturers to perform because it is very expensive. This paper presents a steel plate surface defect detection system with industrial-grade detection performance by training a small amount of steel plate surface images consisting of labeled and non-labeled data. To overcome the problem of lack of training data, we propose two data augmentation techniques: program-based augmentation, which generates defect images in a geometric way, and generative model-based augmentation, which learns the distribution of labeled data. We also propose a 4-step semi-supervised learning using pseudo labels and consistency training with fixed-size augmentation in order to utilize unlabeled data for training. The proposed technique obtained about 99% defect detection performance for four defect types by using 100 real images including labeled and unlabeled data.

• Design & Manufacturing
• /
• v.15 no.3
• /
• pp.1-6
• /
• 2021

An artificial neural network model based on a deep learning algorithm is known to be more accurate than humans in image classification, but there is still a limit in the sense that there needs to be a lot of training data that can be called big data. Therefore, various techniques are being studied to build an artificial neural network model with high precision, even with small data. The transfer learning technique is assessed as an excellent alternative. As a result, the purpose of this study is to develop an artificial neural network system that can classify burr images of light guide plate products with 99% accuracy using transfer learning technique. Specifically, for the light guide plate product, 150 images of the normal product and the burr were taken at various angles, heights, positions, etc., respectively. Then, after the preprocessing of images such as thresholding and image augmentation, for a total of 3,300 images were generated. 2,970 images were separated for training, while the remaining 330 images were separated for model accuracy testing. For the transfer learning, a base model was developed using the NASNet-Large model that pre-trained 14 million ImageNet data. According to the final model accuracy test, the 99% accuracy in the image classification for training and test images was confirmed. Consequently, based on the results of this study, it is expected to help develop an integrated AI production management system by training not only the burr but also various defective images.

• Journal of the Ergonomics Society of Korea
• /
• v.22 no.1
• /
• pp.69-79
• /
• 2003

This study is plan to show what kinds of problems we have and how we can overcome problems about occupational safety and health education and training system in Korea, Some features which advanced countries had were found and characterized as follows; First, advanced countries realized the importance of safety and health education on the role of preventing accident and they had varieties of training and education programs on the industrial safety and health for employer and worker. Second, there were many safety and health education programs and grants which especially focused to the small businesses. Third, they had also several programs to induce employer and worker's concerning about the safety and health education voluntarily, In conclusion, we introduced some inducing models, and proposed to revise contents of the education programs and to increase participation of other parties in safety and health education market so that could vitalize our safety and health education and training.

• Journal of Korea Society of Industrial Information Systems
• /
• v.23 no.2
• /
• pp.41-52
• /
• 2018

In this study, empirical studies are conducted not only to investigate the effect of the quality of accounting information system on management performance but also to study whether user satisfaction and education training achievement play a mediating role in the relationship between the quality factors of accounting information system and management performance. To this end, a research model is established to verify the mediating effects of user satisfaction and education training achievement as well as the effect of the quality of accounting information system on management performance. We conduct a questionnaire survey on users of accounting information system of companies. The results of the empirical analysis based on the survey data are summarized as follows. First, the quality of information system has a positive effect on educational performance and user satisfaction, but not on business performance. Second, user satisfaction and education training performance play a mediating role in the relationship between the quality of information system and management performance. The results of our research show that it is necessary to educate and to train the employees in order to make the accounting information systeme to contribute management performance and to control continuously the quality of the information system in order to improve user satisfaction.

• Journal of the Society of Disaster Information
• /
• v.2 no.1
• /
• pp.113-127
• /
• 2006

The suggestions that follow are about the Qualitative Improvement of private security Industrial. First, in legal and institutional policy, new establishment by law for private security and more support from government is asked. Moreover, the restructuring or M&A between petty companies and the pricing for security service should be performed. Second, in the structural aspect of private security industry, the professional education center for private security guards should be established and the terms of payment and welfare should be improved to the level above standard. In addition, it should be achieved to change the public to have a new and correct understanding of private security and develope the specialized parts suited to the characteristic and ability of each companies. Third, the construction of operating system for private security service should be achieved; recruit system for competent security guards, marketing strategy and enforcement system, widely known confidence to client, normal training system for security guards and post management system for client. This is also to be suited to the characteristic of each companies.

• Journal of Engineering Education Research
• /
• v.5 no.2
• /
• pp.54-62
• /
• 2002

The study reviewed the properness of a present job field with 26 items so that the classification system of the national technical qualification can meet the needs of the industrial structure and the technical changes of the 21st century. In addition, it aimed not only to improve the efficiency and effectiveness of the management and operation of the national technical qualification system but to design a new classification system of the job field to stimulate qualification holders' employment. For the purpose of this study, materials and data relevant to the national technical qualification system were comprehensively collected and analyzed. Besides, a new job field with 13 job items was proposed in this study on a basis of the collective advice of experts on the properness of the job field in the national technical qualification system.

• Journal of the Korea Safety Management & Science
• /
• v.3 no.2
• /
• pp.65-71
• /
• 2001

Comparing the rate of accident in 1999 with 1998, that has been increased about 0.06%. The 56 percentage of industrial accident had been occurred by new employer, who has been worked less than 1 year, in 1999, therefore safety training and education is required for those new employer. This study is present the analysis and investigation of safety training and educational situation and consider a counterplan.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.4
• /
• pp.867-875
• /
• 2019

Security awareness and training are becoming more important as cyber security incidents tend to increase in industrial control systems, including nuclear power plants. For effective cyber security awareness and training for the personnel who manage and operate the target facility, a TEST-BED is required that can analyze the impact of cyber attacks from the sensor level to the operation status of the nuclear power plant. In this paper, we have developed an HIL system for nuclear power plant cyber security training. It includes nuclear power plant status simulations and specific system status simulation together with physical devices. This research result will be used for the specialized cyber security training program for Korean nuclear facilities.

• International Journal of Quality Innovation
• /
• v.7 no.3
• /
• pp.58-69
• /
• 2006

This paper presents neural network-based recognition system for automatic light emitting diode (LED) inspection. The back-propagation neural network (BPNN) is proposed and tested. The current-voltage (I-V) characteristic data of LED from the inspection process is used for the network training and testing. This study selects 300 random samples as network training and employs 100 samples as network testing. The experimental results show that if the classification work is done well, the accuracy of recognition is 100%, and the testing speed of the proposed recognition system is almost one half faster than the traditional inspection system does. The proposed neural-network approach is successfully demonstrated by real data sets and can be effectively developed as a recognition system for a practical application purpose.

• The Journal of Industrial Distribution & Business
• /
• v.6 no.3
• /
• pp.9-17
• /
• 2015

Purpose - The study's aim is to assess performance quality delivered through a performance management system. The case study first measures the effectiveness of the present performance management system of Bhushan Steel Ltd. (BSL), and then secondarily analyzes the employee training impact. Research design, data, and methodology - A case study research approach using a survey was followed. The final sample size was 50, with a simple random sampling technique used. Primary data came from the company itself and the secondary from books, the Internet, journals, etc. Results - BSL employee appraisals are used for many developmental aspects such as competence in the present job, improvements, performance development, and training needs. The results indicate that individual performance can improve if employees perceive the system as more transparent (i.e., not so confidential). Conclusions - Overall, the BSL employees think that the company appraisal system is useful; however, to make it better the company has to improve many aspects including appraisals based on supervisor observations. The company could use a self-appraisal and a 360-appraisal approach to improve such assessments.