• Journal of Korea Society of Digital Industry and Information Management
• /
• v.19 no.3
• /
• pp.245-251
• /
• 2023

Facial expression recognition plays a significant role in understanding human emotional states. With the advancement of AI and computer vision technologies, extensive research has been conducted in various fields, including improving customer service, medical diagnosis, and assessing learners' understanding in education. In this study, we develop a model that can infer emotions in real-time from a webcam using transfer learning with TensorFlow.js and MobileNet. While existing studies focus on achieving high accuracy using deep learning models, these models often require substantial resources due to their complex structure and computational demands. Consequently, there is a growing interest in developing lightweight deep learning models and transfer learning methods for restricted environments such as web browsers and edge devices. By employing MobileNet as the base model and performing transfer learning, our study develops a deep learning transfer model utilizing JavaScript-based TensorFlow.js, which can predict emotions in real-time using facial input from a webcam. This transfer model provides a foundation for implementing facial expression recognition in resource-constrained environments such as web and mobile applications, enabling its application in various industries.

• Transactions of Materials Processing
• /
• v.32 no.4
• /
• pp.221-227
• /
• 2023

While tube bending is a conventional forming technique, it is still used to make curved products for load-bearing members or aesthetically pleasing parts in various manufacturing industries such as automotive, aerospace, and others. Whole or local deformation of the final product such as springback, distortion, or local buckling are of interest in metal forming or precision manufacturing. In this paper, the factors affecting the cross-sectional deformation are explored. A 5-axis stretch bending machine was used for two-dimensional bending with extruded AA6082-T4 rectangular tubes. Three different bending sequences were employed: stretch before bending, stretch after bending, simultaneous bending and stretch. Furthermore, by considering both the stretch and bending angle, cross-sectional deformation was also analyzed. It was observed that employing stretch bending techniques can effectively reduce cross-sectional deformation and contribute to overall quality enhancement. Through this study, it was revealed that these factors have an impact on the cross-sectional deformation of the tubes.

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.1 no.2
• /
• pp.181-191
• /
• 1991

The present study determined the overall noise level and the distribution of sound pressure level over audible frequency range of noise produced at various work sites. Work-related noise greater than 80dBA produced from 98 separate work sites at 37 manufacturing companies and machine shops were analysed for the overall sound level (dBA) and frequency distribution. In addition, to determine the possible hearing loss related to work site noise, a hearing test was also conducted on 1,374 workers in these work sites. The results of the study were as follows ; 1. Of the total 98 work sites, 57 work sites(58.2%) produced noise exceeding threshold limit value (${\geq}90dBA$) set by the Ministry 01 Labor. In terms of different manufacturing industries the proportion of work sites which exceeded 90dBA was the highest for the cut-stone products industry with 6/6 work sites and lowest for the commercial printing industry with 1/13 work sites. 2. The percentage of workers who were exposed to noise greater than 90dBA was 19.8% (1,040 workers) 01 the total 5,261 workers. In terms of different industries, cut-stone products industry had the most workers exposed to noise exceeding 90dBA with 82.8%, textile bleaching and dyeing industry was next at 30.6% followed by fabricated metal products industry with 27.9%, plastic products manufacturing industry had the lowest percentage of workers exposed to 90dBA exceeding noise with 4.5%. 3. There was a statistically significant correlation between the frequency of noise-induced hearing loss and the percentage of workers exposed to noise exceeding 90dBA (P<0.05). 4. The frequency analysis of noise produced at the 98 work sites revealed that 44 work sites (44.9%) had the maximum sound pressure level at high-frequencies greater than 2KHz. In addition, significantly higher sound pressure level was detected at the high-frequencies at 90dBA exceeding work sites as compared to below 90dBA work sites (P<0.01). 5. The differences in sound level meter's A-and C-weighted sound pressure levels were analysed by frequencies. Of the 28 work sites which showed 0-1 dB difference in the two weighted sound levels, 20 work sites (71.4%) had significantly higher sound pressure levels at high-frequencies greater than 2KHz (P<0.01). Furthermore, there was a tendency for higher sound pressure levels to occur in the high-frequency range as the differences in the two weighted sound levels decreased.

• Journal of Korea Port Economic Association
• /
• v.25 no.4
• /
• pp.107-130
• /
• 2009

Future of Busan New Port may depend even on the efficient use of the port hinterland. Accordingly, selection of which industry according to which standard in the port hinterland is another task. In order to solve this problem, it analyzed the structure in international division of labor with China and Japan, which are possessing considerable portion in the trading volume with our country, and the export-import structure of Busan Port against China and Japan, by using RCA index and GL index as well as export-import results. In addition to this, the proper industry was selected on the basis of 10 strategic industries for development in Busan. According to the analytical results, the industries, which will be induced in the hinterland of Busan New Port, include textile clothing, pulp printing matter, jewelry, basic metal nonmetallic product, machine lectric product, automobile, shipbuilding, optics accurate machinery medical treatment musical instrument, nano material, fuel battery, aerospace and intelligent robot.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.6
• /
• pp.9-19
• /
• 2016

In finance literature, stock liquidity showing how stocks can be cashed out in the market has received rich attentions from both academicians and practitioners. The reasons are plenty. First, it is known that stock liquidity affects significantly asset pricing. Second, macroeconomic announcements influence liquidity in the stock market. Therefore, stock liquidity itself affects investors' decision and managers' decision as well. Though there exist a great deal of literature about stock liquidity in finance literature, it is quite clear that there are no studies attempting to investigate the stock liquidity issue as one of decision making problems. In finance literature, most of stock liquidity studies had dealt with limited views such as how much it influences stock price, which variables are associated with describing the stock liquidity significantly, etc. However, this paper posits that stock liquidity issue may become a serious decision-making problem, and then be handled by using data mining techniques to estimate its future extent with statistical validity. In this sense, we collected financial data set from a number of manufacturing companies listed in KRX (Korea Exchange) during the period of 2010 to 2013. The reason why we selected dataset from 2010 was to avoid the after-shocks of financial crisis that occurred in 2008. We used Fn-GuidPro system to gather total 5,700 financial data set. Stock liquidity measure was computed by the procedures proposed by Amihud (2002) which is known to show best metrics for showing relationship with daily return. We applied five data mining techniques (or classifiers) such as Bayesian network, support vector machine (SVM), decision tree, neural network, and ensemble method. Bayesian networks include GBN (General Bayesian Network), NBN (Naive BN), TAN (Tree Augmented NBN). Decision tree uses CART and C4.5. Regression result was used as a benchmarking performance. Ensemble method uses two types-integration of two classifiers, and three classifiers. Ensemble method is based on voting for the sake of integrating classifiers. Among the single classifiers, CART showed best performance with 48.2%, compared with 37.18% by regression. Among the ensemble methods, the result from integrating TAN, CART, and SVM was best with 49.25%. Through the additional analysis in individual industries, those relatively stabilized industries like electronic appliances, wholesale & retailing, woods, leather-bags-shoes showed better performance over 50%.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38C no.3
• /
• pp.278-287
• /
• 2013

Modern information technologies continue to provide industries with new and improved methods. With the rapid development of Machine to Machine (M2M) communication, a smart container supply chain management is formed based on high performance sensors, computer vision, Global Positioning System (GPS) satellites, and Globle System for Mobile (GSM) communication. Existing supply chain management has limitation to real time container tracking. This paper focuses on the studies and implementation of real time container chain management with the development of the container identification system and automatic alert system for interrupts and for normal periodical alerts. The concept and methods of smart container modeling are introduced together with the structure explained prior to the implementation of smart container tracking alert system. Firstly, the paper introduces the container code identification and recognition algorithm implemented in visual studio 2010 with Opencv (computer vision library) and Tesseract (OCR engine) for real time operation. Secondly it discusses the current automatic alert system provided for real time container tracking and the limitations of those systems. Finally the paper summarizes the challenges and the possibilities for the future work for real time container tracking solutions with the ubiquitous mobile and satellite network together with the high performance sensors and computer vision. All of those components combine to provide an excellent delivery of supply chain management with outstanding operation and security.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.11
• /
• pp.256-264
• /
• 2020

Industrial Wireless Sensor Networks (IWSNs) is exploited to achieve various objectives such as improving productivity and reducing cost in the diversity of industrial application, and it has requirements such as low-delay and high reliability packet transmission. To accomplish the requirement, the network manager performs graph construction and resource allocation about network topology, and determines the transmission cycle and path of each node in advance. However, this network management scheme cannot treat mobile devices that cause continuous topology changes because graph reconstruction and resource reallocation should be performed as network topology changes. That is, despite the growing need of mobile devices in many industries, existing scheme cannot adequately respond to path failure caused by movement of mobile device and packet loss in the process of path recovery. To solve this problem, a network management scheme is required to prevent packet loss caused by mobile devices. Thus, we analyse the location and movement cycle of mobile devices over time using machine learning for predicting the mobility pattern. In the proposed scheme, the network manager could prevent the problems caused by mobile devices through performing graph construction and resource allocation for the predicted network topology based on the movement pattern. Performance evaluation results show a prediction rate of about 86% compared with actual movement pattern, and a higher packet delivery ratio and a lower resource share compared to existing scheme.

• The Journal of the Acoustical Society of Korea
• /
• v.30 no.8
• /
• pp.446-460
• /
• 2011

The use of induction motors has been recently increasing with automation in aeronautical and automotive industries, and it playes a significant role. This has motivated that many researchers have studied on developing fault detection and classification systems of an induction motor in order to minimize economical damage caused by its fault. With this reason, this paper proposed feature vector extraction methods based on STE (short-time energy)+SVD (singular value decomposition) and DCT (discrete cosine transform)+SVD techniques to early detect and diagnose faults of induction motors, and classified faults of an induction motor into different types of them by using extracted features as inputs of BPNN (back propagation neural network) and multi-layer SVM (support vector machine). When BPNN and multi-lay SVM are used as classifiers for fault classification, there are many settings that affect classification performance: the number of input layers, the number of hidden layers and learning algorithms for BPNN, and standard deviation values of Gaussian radial basis function for multi-layer SVM. Therefore, this paper quantitatively simulated to find appropriate settings for those classifiers yielding higher classification performance than others.

• Proceedings of the KACD Conference
• /
• 2001.11a
• /
• pp.577-577
• /
• 2001

The lastest concepts in bonding are "total etch", in which both enamel and dentin are etched with an acid to remove the smear layers, and "wet dentin" in which the dentin is not blown dry but left moist before application of the bonding primer. Ideally, the application of a bonding agent to tooth structure should be insensitive to minor contamination from oral fluids. Clinically contaminations such as saliva, gingival fluid, blood and handpiece lubricant are often encountered by dentists during preparation of a restoration. The aim of this study was to evaluate the effect of contamination by hem-ostatic agents on shear bond strength of compomer restorations. One hundred and ten extracted human maxillary and mandibular molar teeth were collected. The teeth were cleaned from soft tissue remnant and debris and stored in physiologic solution until they were used. Small flat area on dentin of the buccal surface were wet ground serially with 400, 800 and 1200 abrasive paper on automatic polishing machine. The teeth were randomly divided into 11 groups. Each group was conditioned as follows: Group 1 : Dentin surface was not etched and not contaminated by hemostatic agents. Group2 : Dentin surface was not etched but was contaminated by Astringedent (Ultradent product Inc., Utah, U.S.A.). Group3 : Dentin surface was not etched but was contaminated by Bosmin (Jeil Phann, Korea.). Group4 : Dentin surface was not etched but was contaminated by Epri-dent (Epr Industries, NJ, U.S.A.). Group5: Dentin surface was etched and not contaminated by hemostatic agents. Group 6 : Dentin surface was etched and contaminated by Astringedent. Group7 : Dentin surface was etched and contaminated by Bosmin. Group8 : Dentin surface was etched and contaminated by Epri-dent. Group9 : Dentin surface was contaminated by Astringedent. The contaminated surface was rinsed by water and dried by compressed air. Group10 : Dentin surface was contaminated by Bosmin. The contaminated surface was rinsed by water aud dried by compresfed air. Group 11 : Dentin surface was contaminated by Epri-dent. The contaminated surface was rinsed by water and dried by compresfed air. After surface conditioning, F2000 was applicated on the conditoned dentin surface. The teeth were thermocycled in distilled water at $5^{\circ}C\;and\;55^{\circ}C$ for 1000 cycles. The samples were placed on the binder with the bonded compomer-dentin interface parallel to the lmife-edge shearing rod of the Universal testing machine(Zwick 020, Germany) running at a cross head speed of 1.0mmimin. There were no significant differences in shear bond strength between groups 1 and group 3 and 4, but group 2 showed significant decrease in shear bond strength compared with group 1. There were no significant differences in shear bond strength between group 5 and group 7 and 8, but group 6 showed significant decrease in shear bond strength compared with group 5. There were no significant differences in shear bond strength between group 5 and group 9, 10 and 11.

• Journal of Internet Computing and Services
• /
• v.14 no.5
• /
• pp.119-128
• /
• 2013

As ICT technology becomes advanced, the importance of future internet is emphasized and in part of that, M2M (Machine-to Machine communications) is magnified in terms of importance and usage in public and private sector. M2M is emerging as a next generation strategic industry but there is no existing analyzed data or market classification, so it disrupts establishing policies on the M2M industry. As the technology is progressing, the evolution from M2M to IoT (Internet of Things) has started and many countries actively try to find technological trend through market analysis in order to develop new growth engine. Therefore, in order to strengthen competitiveness, we should secure differentiated capabilities in industry and service. This article examines Korea's domestic market and international market trends in IoT and analyses the economic impact of the IoT industry using quantitative methodology and evaluates relations between the IoT industry and other relevant industries. As a result, the effect of IoT industry on production inducement is KRW474.6 billion; the effect on value-added inducement is KRW314.7 billion; and it is measured that 3,628 jobs will be created by the IoT industry.