• Journal of Broadcast Engineering
• /
• v.23 no.3
• /
• pp.383-394
• /
• 2018

In this paper, we investigate image and video compression techniques based on deep learning which are actively studied recently. The deep learning based image compression technique inputs an image to be compressed in the deep neural network and extracts the latent vector recurrently or all at once and encodes it. In order to increase the image compression efficiency, the neural network is learned so that the encoded latent vector can be expressed with fewer bits while the quality of the reconstructed image is enhanced. These techniques can produce images of superior quality, especially at low bit rates compared to conventional image compression techniques. On the other hand, deep learning based video compression technology takes an approach to improve performance of the coding tools employed for existing video codecs rather than directly input and process the video to be compressed. The deep neural network technologies introduced in this paper replace the in-loop filter of the latest video codec or are used as an additional post-processing filter to improve the compression efficiency by improving the quality of the reconstructed image. Likewise, deep neural network techniques applied to intra prediction and encoding are used together with the existing intra prediction tool to improve the compression efficiency by increasing the prediction accuracy or adding a new intra coding process.

• Journal of the Korean Society for information Management
• /
• v.33 no.4
• /
• pp.125-157
• /
• 2016

As the domestic companies which have made inroads into foreign markets have more lawsuits, these companies' demands for responding to E-Discovery are also increasing. E-Discovery, derived from Anglo-American law, is the system to find electronic evidences related to lawsuits among scattered electronic data within limited time, to review them as evidences, and to submit them. It is not difficult to find, select, review, and submit evidences within limited time given the reality that the domestic companies do not manage their records even though lots of electronic records are produced everyday. To reduce items to be reviewed and proceed the process efficiently is one of the most important tasks to win a lawsuit. The Predictive Coding is a computer assisted review instrument used in reviewing process of E-Discovery, which is to help companies review their own electronic data using mechanical learning. Predictive Coding is more efficient than the previous computer assister review tools and has a merit to select electronic data related to lawsuit. Through companies' selection of efficient computer assisted review instrument and continuous records management, it is expected that time and cost for reviewing will be saved. Therefore, in for companies to respond to E-Discovery, it is required to seek the most effective method through introduction of the professional Predictive Coding solution and Business records management with consideration of time and cost.

• Food Science and Preservation
• /
• v.13 no.4
• /
• pp.438-444
• /
• 2006

This study was carried out to obtain fundamental data such as peeling efficiency and quality of potatoes and sweet potatoes peeled by hand, machine and alkali(NaOH). The weight loss by peeling was influenced by peeling methods. Weight losses by rotational brushing-type peeler showed the lowest value, 7.9% in potato, and 7.3% in sweet potato. Any significant differences in moisture content were not found in potatoes and sweet potatoes by peeling methods. The pH of potatoes and sweet potatoes just after peeling were 5.8-6.8 and 6.23-6.63, and decreased somewhat until 3 hrs after peeling. Hardness of potatoes and sweet potatoes peeled by hand with fruit knife were better than that of others. Depending upon the peeling method used the color and color differences undergo some changes in their color and browning. Color difference value of peeled potatoes by hand with a technical tools, and by mechanical peeler such as rotational cutting-type peeler and rotational brushing-type peeler showed just slightly. In particular, changes of rotor differences value of potatoes and sweet potatoes peeled by dipping with 10% NaOH solution at $100^{\circ}C$ was the highest in the samples peeled by NaOH.

• Proceedings of the Korean Vacuum Society Conference
• /
• 1999.07a
• /
• pp.211-211
• /
• 1999

Titanium nitride (TiN) thin films have useful properties including high hardness, good electrical conductivity, high melting point, and chemical inertness. The applications have included wear-resistant hard coatings on machine tools and bearings, decorative coating making use of the golden color, thermal control coatings for widows, and erosion resistant coatings for spacecraft plasma probes. For all these applications as feature sizes shrink and aspect ratios grow, the issue of good step coverage becomes increasingly important. It is therefore essential to manufacture conformal coatings of TiN. The growth of TiN thin films by chemical vapor deposition (CVD) is of great interest for achieving conformal deposition. The most widely used precursor for TiN is TiCl4 and NH3. However, chlorine impurity in the as-grown films and relatively high deposition temperature (>$600^{\circ}C$) are considered major drawbacks from actual device fabrication. To overcome these problems, recently, MOCVD processes including plasma assisted have been suggested. In this study, therefore, we have doposited Ti(C, N) thin films on Si(100) and D2 steel substrates in the temperature range of 150-30$0^{\circ}C$ using tetrakis diethylamido titanium (TDEAT) and titanium isopropoxide (TIP) by pulsed DC plamsa enhanced metal-organic chemical vapor deposition (PEMOCVD) method. Polycrystalline Ti(C, N) thin films were successfully grown on either D2 steel or Si(100) surfaces at temperature as low as 15$0^{\circ}C$. Compositions of the as-grown films were determined with XPS and RBS. From XPS analysis, thin films of Ti(C, N) with low oxygen concentration were obtained. RBS data were also confirmed the changes of stoichiometry and microhardness of our films. Radical formation and ionization behaviors in plasma are analyzed by optical emission spectroscopy (OES) at various pulsed bias and gases conditions. H2 and He+H2 gases are used as carrier gases to compare plasma parameter and the effect of N2 and NH3 gases as reactive gas is also evaluated in reduction of C content of the films. In this study, we fond that He and H2 mixture gas is very effective in enhancing ionization of radicals, especially N resulting is high hardness. The higher hardness of film is obtained to be ca. 1700 HK 0.01 but it depends on gas species and bias voltage. The proper process is evident for H and N2 gas atmosphere and bias voltage of 600V. However, NH3 gas highly reduces formation of CN radical, thereby decreasing C content of Ti(C, N) thin films in a great deal. Compared to PVD TiN films, the Ti(C, N) film grown by PEMOCVD has very good conformability; the step coverage exceeds 85% with an aspect ratio of more than 3.

• Journal of Broadcast Engineering
• /
• v.25 no.4
• /
• pp.487-496
• /
• 2020

For the purpose of next generation technology for robot perfomances, a RAoRA (Robot Actor on Robot Arm) structure was proposed using a robot arm joined with a humanoid robot actor. Mechanical analysis, machine design and fabrication were performed for motions combined with the robot arm and the humanoid robot actor. Kinematical analysis for 3D model, spline interpolation of positions, motion control algorithm and control devices were developed for movements of the robot actor. Preliminary visualization, simulation tools and integrated operation of consoles were constructed for the non-professionals to produce intuitive and safe contents. Air walk was applied to test the developed platform. The air walk is a natural walk close to a floor or slow ascension to the air. The RAoRA also executed a performance with 5 minute-running time. Finally, the proposed platform of robot performance presented intensive and live motions which was impossible in conventional robot performances.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.12
• /
• pp.182-189
• /
• 2019

3D printing technology is changing the paradigm of consumer-oriented design in supplier-oriented mass production. 3D printing technology in construction is expected to be able to replace existing wet methods along with modular construction. Recently, a number of cases of building construction using 3D printing using mortar-based materials have been announced in many regions, including North America, Europe, and Asia. In this study, we developed a design support tool with a slicing function to output 3D modeling for architecture for a 3D printing machine. We analyzed the process and the function of slicing programs that are commercially available. Seven slicing functions required for the architectural field were derived by analyzing cases, expert reviews, and related literature. The derived functions were extended from the slicing functions to develop the design support tools. Detailed algorithms and processes need to be developed for future derived functions.

• Journal of Intelligence and Information Systems
• /
• v.16 no.4
• /
• pp.173-193
• /
• 2010

The World Wide Web is transitioning from being a mere collection of documents that contain useful information toward providing a collection of services that perform useful tasks. The emerging Web service technology has been envisioned as the next technological wave and is expected to play an important role in this recent transformation of the Web. By providing interoperable interface standards for application-to-application communication, Web services can be combined with component-based software development to promote application interaction and integration within and across enterprises. To make Web services for service-oriented computing operational, it is important that Web services repositories not only be well-structured but also provide efficient tools for an environment supporting reusable software components for both service providers and consumers. As the potential of Web services for service-oriented computing is becoming widely recognized, the demand for an integrated framework that facilitates service discovery and publishing is concomitantly growing. In our research, we propose a framework that facilitates Web service discovery and publishing by combining clustering techniques and leveraging the semantics of the XML-based service specification in WSDL files. We believe that this is one of the first attempts at applying unsupervised artificial neural network-based machine-learning techniques in the Web service domain. We have developed a Web service discovery tool based on the proposed approach using an unsupervised artificial neural network and empirically evaluated the proposed approach and tool using real Web service descriptions drawn from operational Web services repositories. We believe that both service providers and consumers in a service-oriented computing environment can benefit from our Web service discovery approach.

• The KIPS Transactions:PartD
• /
• v.14D no.2
• /
• pp.265-272
• /
• 2007

In this paper, we implement an automated composition system of web services using hybrid techniques that merge the benefit of BPEL techniques, with the advantage of OWL-S, BPEL techniques have practical capabilities that fulfil the needs of the business environment such as fault handling and transaction management. However, the main shortcoming of these techniques is the static composition approach, where the service selection and flow management are done a priori and manually. In contrast, OWL-S techniques use ontologies to provide a mechanism to describe the web services functionality in machine-understandable form, making it possible to discover, and integrate web services automatically. This allows for the dynamic integration of compatible web services, possibly discovered at run time, into the composition schema. However, the development of these approaches is still in its infancy and has been largely detached from the BPEL composition effort. In this work, we describe the design of the SemanticBPEL architecture that is a hybrid system of BPEL4WS and OWL-S, and propose algorithms for web service search and integration. In particular, the SemanticBPEL has been implemented based on the open source tools. The proposed system is compared with existing BPEL systems by functional analysis. These comparisions show that our system outperforms existing systems.

• Journal of the Korean Chemical Society
• /
• v.53 no.6
• /
• pp.653-662
• /
• 2009

Developing effective tools for predicting absorption, distribution, metabolism, excretion properties and toxicity (ADME/T) of new chemical entities in the early stage of drug design is one of the most important tasks in drug discovery and development today. As one of these attempts, support vector machines (SVM) has recently been exploited for the prediction of ADME/T related properties. However, two problems in SVM modeling, i.e. feature selection and parameters setting, are still far from solved. The two problems have been shown to be crucial to the efficiency and accuracy of SVM classification. In particular, the feature selection and optimal SVM parameters setting influence each other, which indicates that they should be dealt with simultaneously. In this account, we present an integrated practical solution, in which genetic-based algorithm (GA) is used for feature selection and grid search (GS) method for parameters optimization. hERG ion-channel inhibitor classification models of ADME/T related properties has been built for assessing and testing the proposed GA-GS-SVM. We generated 6 different models that are 3 different single models and 3 different ensemble models using training set - 1891 compounds and validated with external test set - 175 compounds. We compared single model with ensemble model to solve data imbalance problems. It was able to improve accuracy of prediction to use ensemble model.

• 대한공업교육학회지
• /
• v.41 no.1
• /
• pp.1-21
• /
• 2016

The purpose of this study is to analyze job of die and mold makers for apprenticeship education of specialized high school using DACUM method. The contents of this study were to identify the duties, specific tasks performed and the level of importance, difficulty, frequency and entry level on each task. This study also make out a DACUM chart of die and mold makers for apprenticeship education of specialized high school. The DACUM committee, which consisted of one facilitator, nine die and mold experts, one recorder and one coordinator, was established to conduct DACUM. The conclusions are as follows; First, this study defines a die and mold makers as 'a person who make it possible to manufacture/assemble/inspect die and mold parts using various machine tool'. Second, duties in job of die and mold makers for apprenticeship education were total 12 and total tasks in job of die and mold makers for apprenticeship education were 86. Third, this study determine the level of importance, difficulty, frequency and entry level about each task. Finally, this study make out a DACUM chart of die and mold makers for apprenticeship education of specialized high school based on the results of DACUM job analysis. And knowledge, skills, tools, and positive behaviors, future trends/concerns about die and mold makers for apprenticeship education of specialized high school were presented.