• The Transactions of the Korea Information Processing Society
• /
• v.3 no.6
• /
• pp.1598-1606
• /
• 1996

In order to achieve high accuracy of off-line character recognition(OCR) systems, the recognized text must be processed through a post-processing stage using contextual information. In this paper, we reclassify Korean word classes in terms of OCR word correction. And we collect combinations of Korean particles(approximately 900) linguistic verbal from(around 800). We aggregate 9 Korean irregular verbal phrases defined from a Korean linguistic point of view. Using these Korean word information and a Head-tail method, we can correct misrecognized words. A Korean character recognizer demonstrates 93.7% correct character recognition without a post-processing stage. The entire recognition rate of our system with a post-processing stage exceeds 97% correct character recognition.

• Journal of the Korea Institute of Building Construction
• /
• v.9 no.4
• /
• pp.131-137
• /
• 2009

The rate of recycling of waste concrete, which represents the majority of construction-related waste, is increasing. However, a general recognition of the inferior qualify of recycled aggregates and their lower grade of compressive strength, bending strength, shear strength, frost resistance and ductility make the application of recycled aggregates to structures insufficient. Therefore, this study conducted material and member experiments by adding steel fiber for the purpose of improving the properties of recycled aggregate concrete. To synthesize the experimental results, it was found that specimens with a 30% steel fiber admixture had levels of compressive strength, tensile strength and frost resistance that were equivalent to or higher than the standard specimen, and that concrete that had a 30% replacement of recycled aggregates with steel fiber was suitable for application to actual structures.

• Journal of the Korean Society of Hazard Mitigation
• /
• v.3 no.4 s.11
• /
• pp.119-129
• /
• 2003

Concrete and steel are essential structural materials in the construction. But, concrete, different from steel, consists of many materials and is affected by many factors such as properties of materials, site environmental situations, and skill of constructors. Concrete have two kinds of properties, immediately knowing properties such as slump, air contents and time dependent one like strength. Therefore, concrete mixes depend on experiences of experts. However, at point of time using High Performance Concrete, new method is wanted because of more ingredients like mineral and chemical admixtures and lack of data. Artificial Neural Networks(ANN) are a mimic models of human brain to solve a complex nonlinear problem. They are powerful pattern recognizers and classifiers, also their computing abilities have been proven in the fields of prediction, estimation and pattern recognition. Here, among them, the back propagation network and radial basis function network ate used. Compositions of high-performance concrete mixes are eight components(water, cement, fine aggregate, coarse aggregate, fly ash, silica fume, superplasticizer and air-entrainer). Compressive strength, slump, and air contents are measured. The results show that neural networks are proper tools to minimize the uncertainties of the design of concrete mixtures.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.11
• /
• pp.5427-5445
• /
• 2019

While deep neural networks have achieved remarkable performance in representation learning, a huge amount of labeled training data are usually required by supervised deep models such as convolutional neural networks. In this paper, we propose a new representation learning method, namely generative adversarial networks (GAN) based bagging deep convolutional autoencoders (GAN-BDCAE), which can map data to diverse hierarchical representations in an unsupervised fashion. To boost the size of training data, to train deep model and to aggregate diverse learning machines are the three principal avenues towards increasing the capabilities of representation learning of neural networks. We focus on combining those three techniques. To this aim, we adopt GAN for realistic unlabeled sample generation and bagging deep convolutional autoencoders (BDCAE) for robust feature learning. The proposed method improves the discriminative ability of learned feature embedding for solving subsequent pattern recognition problems. We evaluate our approach on three standard benchmarks and demonstrate the superiority of the proposed method compared to traditional unsupervised learning methods.

• Journal of Fashion Business
• /
• v.12 no.3
• /
• pp.16-30
• /
• 2008

The purpose of this study is to inquire into the necessity and problem of the study on beauty aesthetics, and the direction and method of its study theoretically by applying the concept of aesthetics to the beauty art as an attempt to build up aesthetic concepts of beauty art. As results of inquiry, the study on beauty aesthetics is essential for the adaptation to demands for the creativity of beauty art. As for problem of study on beauty aesthetics, it is necessary to conduct an aggregate study at various angles since the aesthetic value of beauty art (hairstyle) is realized by interaction of the social cultural value or practical value and pure artistic value, and the unique cosmetic beauty of any era or nation group individual is created on the basis of hair designers and individual spirit, emotion and taste, etc. and its perceptual aesthetic enjoyment is accomplished variously. As for the direction and method of study on beauty aesthetics, the quest with abstract speculative self-examination into the beauty having its subjective meaning must be accomplished. This thesis would like to contribute to the building up of beauty industry field as beauty arts based on the aesthetics by successive coming forward of personnel for beauty who combine the aesthetic abilities meeting the diversified requirements of modern society and by improving the recognition of persons recognizing the beauty industry field as a technical field only.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.8
• /
• pp.1996-2015
• /
• 2023

High-definition (HD) maps can provide precise road information that enables an autonomous driving system to effectively navigate a vehicle. Recent research has focused on leveraging semantic segmentation to achieve automatic annotation of HD maps. However, the existing methods suffer from low recognition accuracy in automatic driving scenarios, leading to inefficient annotation processes. In this paper, we propose a novel semantic segmentation method for automatic HD map annotation. Our approach introduces a new encoder, known as the convolutional transformer hybrid encoder, to enhance the model's feature extraction capabilities. Additionally, we propose a multi-level fusion module that enables the model to aggregate different levels of detail and semantic information. Furthermore, we present a novel decoupled boundary joint decoder to improve the model's ability to handle the boundary between categories. To evaluate our method, we conducted experiments using the Bird's Eye View point cloud images dataset and Cityscapes dataset. Comparative analysis against stateof-the-art methods demonstrates that our model achieves the highest performance. Specifically, our model achieves an mIoU of 56.26%, surpassing the results of SegFormer with an mIoU of 1.47%. This innovative promises to significantly enhance the efficiency of HD map automatic annotation.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.19 no.5
• /
• pp.22-29
• /
• 2015

Modern structures is the tendency of being increasingly taller and larger. The concrete with large weight has the disadvantage of increasing the weight on the structure. therefore, the method of carrying out the weight saving of the concrete is required. one of such method is to use a lightweight aggregate. However, studies on structural lightweight concrete, lacking for the recognition of the lightweight concrete, so also is lacking. therefore it is necessary to study on the physical characteristic value of the lightweight concrete. In this study, in order to investigate the tensile properties of lightweight concrete, Crack mouth opening displacement (CMOD) experiments were carried out. the fracture energy of the lightweight concrete subjected to inverse analysis were derived from the CMOD experimental results.

• The Korean Journal of Archival Studies
• /
• no.10
• /
• pp.145-194
• /
• 2004

Church archives are the evidential instruments to remember church activity and important information aggregate which has administrative, legal, financial, historical, faithful value as the collective memory of church community. So it must be managed necessarily and the management orders are based on the Bible. The western churches which have a correct understanding about the importance of church records and management order have taken multilateral endeavor to create, manage church archives systematically. On the other hand, korean churches don't have the records management systems. Therefore, Records created in individual church are mostly managed unsystematically and exist as 'backlogs', finally, they are destructed without reasonable formalities. In those problems, the purpose of this study is to offer the way of records classification and disposition instrument with recognition that records management should be done from the time of creation or previous to it. As a concrete device for them, I tried to embody the function-based classification method and disposal schedule. I prefer the function-based classification and disposal schedule to the organization and function-based classification to present stable classification and disposal schedule, as we can say the best feature of the modern organization is multilateral and also churches have same aspect. For this study, I applied DIRKS(Designing and Implementing Recordkeeping Systems) manual which National Archives of Australia provide and guidelines in ICA/IRMT series to construct the theory of the function-based classification in individual churches. Through them, it was possible to present a model for preliminary investigation, analysis of business activity, records survey, disposal schedule. And I took an example of 'Myong Sung Presbyterian Church' which belong to 'The Presbyterian church in Korea'. I explained in detail codifying process and results of preliminary investigation in 'Myong Sung Presbyterian Church', analysis of business activity based on it, process of presenting the function-based classification and disposal schedule got from all those steps. For establishing disposal schedule, I planned 'General Disposal Schedule' and 'Agency Disposal Schedule' which categorized 'general function' and 'agency function' of an agency, according to DIRKS in Australia and ICA/IRMT. And for estimation of disposal date I had a thorough grasp of important records category presented in 'Constitution of General Assembly', interview to know the importance of tasks, and added examples of disposal schedule in western church archives. This study has significance that it was intended to embody 'the function-based classification' and 'disposal schedule' suitable for individual church, applying DIRKS in Australia and ICA/IRMT on absence of the theory or example which tried to present the function-based classification and disposal schedule for individual church. Also it is meaningful to present a model that can classify and disposal real records according to the function in individual church which has no recognition or way about records management.

• Journal of Navigation and Port Research
• /
• v.38 no.6
• /
• pp.663-672
• /
• 2014

In Korea and the Philippines, as well as all over the world, with the recognition of the importance of marine ecological resources, the marine protected areas(MPA) have been established and managed to protect and preserve these resources. While the number of marine protected areas for marine ecological resources protection has been increased, there is main problem that the most of MPAs do not achieve their intended management objectives. the effective management. Because of the positive and negative impacts on local communities and fishermen as direct stockholders, there has been ongoing debate on the pros and cons of implementing MPAs. Accordingly, this research conducted a case study of establishing Marine Protected Areas in Guimaras, Philippines because Philippines fisheries code of 1998 (Republic Act 8550), which is enacted to manage, conserve and protect fishery resources, obliged local governments to designate no less than 15% of jurisdictional municipal water as fisheries resource protection areas for a long time. To do this, a dichotomous-choice contingent-valuation survey was conducted in the two municipalities of Guimaras, Philippines to investigate public opinion in debates over MPAs and to estimate willingness to pay (WTP) for MPAs to protect and conserve marine habitats for fishery resources. Because of the expected economic costs by prohibiting fishing activities within the establishing newMPA, 58.7% of respondents thought the costs should be compensated, but 91.4% respondents voted in favor of increasing MPAs for fisheries resources as a protective measure. Finally, with Contingent Valuation Method(CVM), the aggregate mean WTP (375.5ha) of San Lorenzo and Sibunag residents in Guimaras Province, Philippines for establishing the additional MPA in their municipality waters was estimated to $1,046,791. Therefore, these findings could be used as a valuable data for establishing effective management plan of MPAs in Korea.