• International Journal of Contents
• /
• v.8 no.1
• /
• pp.39-46
• /
• 2012

Character segmentation is a preprocessing step in many offline handwriting recognition systems. In this paper, Chinese characters are categorized into seven different structures. In each structure, the character size with the range of variations is estimated considering typical handwritten samples. The component removal and merge criteria are presented to remove punctuation symbols or to merge small components which are part of a character. Finally, the criteria for segmenting the adjacent characters concerning each other or overlapped are proposed.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2004.03b
• /
• pp.600-605
• /
• 2004

The road construction with horizontal expansion of country using and augmentation of traffic demand is advanced actively and it accompanies hereupon, the above of 70% of the country is formed at the mountain in our country where the hazard cut slope has been created. In this study, It is prepared a effective management countermeasure of cut slope introduced priority investigation decision method against hazard cut slope which is influenced by abnormally rainfall by an unusual change in the weather such as a guerilla rainfall character. In meaning link, It was executed collapse cause by failure character analysis in the cut slope which has failed for the last five years and it is prepared the hazard grade criterion from E to A grade according to collapse cause. It is decided that a maintenance management grade by the hazard grade classification criterion of cut slope. So It is possible to hazard cut slope. It is established failure protection counter countermeasure by effective maintenance management through the hazard grade c1assification criterion and it will be able to dispose to advanced nation level like Hong Kong and Japanese.

• Journal of Multimedia Information System
• /
• v.6 no.4
• /
• pp.225-234
• /
• 2019

Deep learning has become increasingly popular in both academic and industrial areas nowadays. Various domains including pattern recognition, Computer vision have witnessed the great power of deep neural networks. However, current studies on deep learning mainly focus on quality data sets with balanced class labels, while training on bad and imbalanced data set have been providing great challenges for classification tasks. We propose in this paper a method of data analysis-based data reduction techniques for selecting good and diversity data samples from a large dataset for a deep learning model. Furthermore, data sampling techniques could be applied to decrease the large size of raw data by retrieving its useful knowledge as representatives. Therefore, instead of dealing with large size of raw data, we can use some data reduction techniques to sample data without losing important information. We group PCB characters in classes and train deep learning on the ResNet56 v2 and SENet model in order to improve the classification performance of optical character recognition (OCR) character classifier.

• 중국학논총
• /
• no.64
• /
• pp.53-73
• /
• 2019

In the teaching of Chinese characters, making full use of the phonetic function of phonetic symbols can help learners improve their learning efficiency. The research on the characteristics of phonetic symbols and the rules of their construction is the premise of teaching Chinese characters with phonetic symbols. The phonetic symbols that can accurately prompt the pronunciation of the whole word and the homophone characters that they constitute provide the applicable materials for the teaching of Chinese characters. The split method simply and intuitively reflects the internal relationship among shape, sound and meaning in pictophonetic characters. "The analogy method of homophonic character group" and "the converse method of homophonic character group" are the combination of the function of the sound prompt and the characteristics of the analogy and induction of homophonic character, which can not only help students save the time of memorizing the sound, but also effectively increase the amount of literacy. The quantitative analysis of phonetic symbols and homophone symbols is of great significance to the classification of Chinese characters and the improvement of textbook editing.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.6
• /
• pp.2751-2756
• /
• 2012

Selection of training set influences the classification accuracy in supervised classification using satellite imagery. Generally, if pure pixels which character of training set is clear were selected, whole accuracy is high while if mixed pixels were selected, accuracy is decreased because of low-resolution imagery or unclear distinguishment. However, it is too difficult to choose the pure pixels as training set actually. Accordingly, this study should be suggested the suitable classification method in case of mixed pixels choice. To achieve this, a few pure pixels were chosen as training set and classification accuracy was calculated which was compared with classification result using an equal number of mixed pixels. As a result, accuracy of SVM was the highest among the classification method using mixed pixels and it was a relatively small difference with the result of classification using pure pixels. Therefore, imagery classification using SVM is most suitable in the mixed area of construction and green because it is high possibility to choose mixed pixels as training set.

• Proceeding of Spring/Autumn Annual Conference of KHA
• /
• 2003.11a
• /
• pp.277-282
• /
• 2003

The purpose of this study is to investigate the architectural visual identity, and scheme of conceptional extent of architectural visual identity through literature research and sample analysis by design factors for apartment design. Korea National Housing Corporation's Mapo apartment put first caterer which is continued about 30 years to solve shortage of houses by beginning and the impact of IMF and the price decontrol on second half in the 1990s have changed the housing status to commercial goods. Construction companies have introduced brand concept to satisfy consumer's appetite in production of house goods including apartment under market situation. Importance of visual identity in apartment design occupies has risen in heftily. However most companies are in condition of stay on two-dimensional application that uses symbol, logotype, and character assistance. Several high position construction companies are trying architectural visual identity, but those are applied in partial element and those are not going forward to systematic classification and synthetic visual identity plan within peculiarity of apartment complex. This study is to grasp plan and design tendency to acquire identity of design element of construction field and goods through design examination. In conclusion visual identity in apartment design is architectural visual identity on all of 2 dimensions and 3 dimension for application. Architectural visual identity is a language of form that prescribe identity about external appearance in combination with brand image.

• KIEAE Journal
• /
• v.12 no.6
• /
• pp.77-84
• /
• 2012

The current architectural design of unit modular has been based on 2D of CAD program, so unit modular character which needs unit information management, as a dried-member system, has no effect on design process. The purpose of this study is We have developed a suitable BIM design process, according to various works of construction, then tried to contribute to supply and activation of the urban-life-housing based on unit modular. The BIM modeling process based on unit modular has been in order of unit combination with preparing manual classification, and, it has been constructed, at construction site, from housing foundation to roof finish by Bottom-up method. At a manufacturing factory, it has been produced in order of 1) grouping materials and parts, 2) fabricating unit boxes, and 3) interference examination of unit boxes, and each order has been classified as housing structure, architecture, plumbing process separately. At a construction site, the fabrication has been done in order of, like as a real housing construction scenario, 1) RC foundation work 2) unit module job-site-fabrication work, 3) roof truss work, 4) plumbing and HVAC work, and 5) housing interior finish work. After modeling process, the interference examination on each work of construction has finally completed modeling. The Unit modular utilizing BIM modeling can make easy housing maintenance through systematic control with preparing manual of unit module information, and securing accurate and speedy construction information. And it will promote design credibility and create maximum effect of unit modular construction method, such as construction period reduction and upgrade of construction quality, etc., through the computer simulation as real as construction environment in cyber space, and with the interfering examination.

• Economic and Environmental Geology
• /
• v.57 no.5
• /
• pp.473-486
• /
• 2024

Boring logs are essential for the evaluation of ground stability in abandoned mine areas, representing geomaterial and subsurface structure information. However, because boring logs are maintained in various analog formats, extracting useful information from them is prone to human error and time-consuming. Therefore, this study develops an algorithm to efficiently manage and analyze boring log data for abandoned mine ground investigation provided in PDF format. For this purpose, the EfficientNet deep learning model was employed to classify the boring logs into five types with a high classification accuracy of 1.00. Then, optical character recognition (OCR) and PDF text extraction techniques were utilized to extract text data from each type of boring log. The OCR technique resulted in many cases of misrecognition of the text data of the boring logs, but the PDF text extraction technique extracted the text with very high accuracy. Subsequently, the structure of the database was established, and the text data of the boring logs were reorganized according to the established schema and written as structured data in the form of a spreadsheet. The results of this study suggest an effective approach for managing boring logs as part of the transition to digital mining, and it is expected that the structured boring log data from legacy data can be readily utilized for machine learning analysis.

• Proceedings of the Korean Institute Of Construction Engineering and Management
• /
• autumn
• /
• pp.155-162
• /
• 2001

Recently, in Korea, a few construction companies have been tried to develop a management system, which is able to integrate schedule and cost. In spite of these attempts, however, advanced management techniques can be hardly applied under the BoQ based management system. In order to improve these problems, many studies have been peformed, but yet could not overcome practical limitations. Besides, the application of historical data is below the level since it is so difficult to accumulate and feed-back historical data under the unique character of construction industry. Consequently, lots of time and effort have being wasted to establish control criteria. The newly generated Information is not systematically managed as well. Therefore, this study suggests Integrated Construction Management System complemented the existing practical problems.

• Journal of architectural history
• /
• v.21 no.1
• /
• pp.99-118
• /
• 2012

This study aims to classify the architectural formation of the Stylobate of and Staircase types of Post-Lintel Construction in Backje style and Silla style and also to figure out their specific elements by periods and transformation characteristics. The scope of this study was restricted to architectural remains between the Three Kingdom period and the Goryeo period. To progress the investigation, the study classified remains in a similar type by period and type, and then and analyzed its character based on formation method and specific factors of subsidiary materials. As a result, its type can be classified as Backje type(I), Silla type(II) and Mixed type(III). Regionally, it is found that Woongjin, Sabi and surroundings in the capital of Backje, Gyeongju and main towns in the capital of Silla, and Gaeseong and main towns in the capital of Goryeo. In particular, type III has the characteristics of type I and II simultaneously, which has one or two layers Jangdaeseok stone between the foundation stone of stylobate and plate stone, and tends to be decorative when Taengjuseok stone is installed. For types of staircase, it could be classified as 6~7C Backje type(I), 7~9C Silla type(II) and 9C Mixed type(III) according to the formation method of Somaetdol (banister of stone stair). And from the 9th century on, decorated Somaetdol stone type(IV) of 1/4 circle shape began to appear. The common feature of stylobate of post-lintel and staircase is that it became simplified, abbreviated and decorative. It seems to be intended to secure simplicity of construction or structural stability.