• Proceedings of the Zoological Society Korea Conference
• /
• 1999.10b
• /
• pp.263.1-263
• /
• 1999

No Abstract, See Full Text

• Proceedings of the Zoological Society Korea Conference
• /
• 2000.10a
• /
• pp.265.2-266
• /
• 2000

No Abstract, See Full Text

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.6 no.2
• /
• pp.181-189
• /
• 2002

In this paper, we designed and implemented Avata construction system using XML(extensible Markup Language) and SVG(Scalable Vector Graphic). The Web character created with Avata(or Web character) construction system are displayed in same (on without damage of image, regardless terminal type and user can modify and change image easily in form that want. Compare with existing Web character system, the Reusability of web character part element Is increased greatly with Avata construction system of this paper. Because SVG is described by text, graphic retrieval is convenient, and applications can use easily SVG document. Also, SVG can create web graphic document dynamically with database because can access easily in all graphic primitives of line, Polygon, text, image etc. As well as web character using study finding, we may develop usable technology to some contents on World Wide Web.

• Proceedings of the IEEK Conference
• /
• 1998.10a
• /
• pp.841-844
• /
• 1998

Korean Text to Sign Language Traslator could be applied to learn letters for both the deaf and hard-of-hearing people, and to have a conversation with normal people. This paper describes some useful dictionaries for developing korean text to sign language translator; Base sign language dictionary, Compound sign language dictionary, and Resemble sign language dictionary. As korean sign language is composed entirely of about 6,000 words, the additional dictionaries are required for matching them to korean written language. We design base sign language dictionary which was composed of basic symbols and moving picture of korean sign language, and propose the definition of compound isng language dictionary which was composed of symbols of base sing language. In addition, resemble sign language dictionary offer sign symbols and letters which is used same meaning in conversation. By using these methods, we could search quickly sign language during korean text to sign language translating process, and save storage space. We could also solve the lack of sign language words by using them, which are appeared on translating process.

• ETRI Journal
• /
• v.32 no.5
• /
• pp.729-739
• /
• 2010

Due to the rapid development of mobile devices equipped with cameras, instant translation of any text seen in any context is possible. Mobile devices can serve as a translation tool by recognizing the texts presented in the captured scenes. Images captured by cameras will embed more external or unwanted effects which need not to be considered in traditional optical character recognition (OCR). In this paper, we segment a text image captured by mobile devices into individual single characters to facilitate OCR kernel processing. Before proceeding with character segmentation, text detection and text line construction need to be performed in advance. A novel character segmentation method which integrates touched character filters is employed on text images captured by cameras. In addition, periphery features are extracted from the segmented images of touched characters and fed as inputs to support vector machines to calculate the confident values. In our experiment, the accuracy rate of the proposed character segmentation system is 94.90%, which demonstrates the effectiveness of the proposed method.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.28 no.1
• /
• pp.92-108
• /
• 2000

The rear garden of the Ch'angdok Palace is considered to be representative of gardens of the 15th century to the end of Choson Dynasty in the early 20th century and is very well preserved. Therefore with its long history, it's been represented in various way. This study tried to interprete the rear garden of the Ch'angdok Palace in terms of 'text through king Jungjo's "Sangrimshipkyog". Jungjo's "Sangrimshipkyng" that describes scenery related to ten place of the rear garden of the Ch'angdok Palace was written when he was a crown prince. And the scenery of ten places can't be separated from their physical looks and those ten places were quite different from now. For that reason, to begin with, this study recomposed those ten places through maps and drawings that revived the current looks and rear garden. And what poems tried to tell and what meanings whey had in the literacy of that age that allowed "poem enchained ten scapes" were studied in the context of co-relationship with space. As a text it is very ancient and questions arise on how to correctly interpret the information it holds. But this interpretation is only one side of the various aspects that the rear garden of the Ch'angdok Palace has compared to its long history. In light of the present age an interpretation can have richer and more varied meanings, creating many new viewpoints. creating many new viewpoints.

• Journal of Information Management
• /
• v.34 no.1
• /
• pp.59-75
• /
• 2003

This study described about development the integrated management system for research reports and construction full text database of KAERI reports. This system enables centralization of various information related to reports by connecting with management information system. The advantages of this study are consistency of database between digital library and MIS, reduction of budget and manpower and increase of research productivity.

• Journal of Information Processing Systems
• /
• v.20 no.2
• /
• pp.215-225
• /
• 2024

With the rapid development of Internet of Things (IoT) and big data technology, a large amount of data will be generated during the operation of related industries. How to classify the generated data accurately has become the core of research on data mining and processing in IoT industry chain. This study constructs a classification model of IoT industry chain based on improved random forest algorithm and text analysis, aiming to achieve efficient and accurate classification of IoT industry chain big data by improving traditional algorithms. The accuracy, precision, recall, and AUC value size of the traditional Random Forest algorithm and the algorithm used in the paper are compared on different datasets. The experimental results show that the algorithm model used in this paper has better performance on different datasets, and the accuracy and recall performance on four datasets are better than the traditional algorithm, and the accuracy performance on two datasets, P-I Diabetes and Loan Default, is better than the random forest model, and its final data classification results are better. Through the construction of this model, we can accurately classify the massive data generated in the IoT industry chain, thus providing more research value for the data mining and processing technology of the IoT industry chain.

• International conference on construction engineering and project management
• /
• 2024.07a
• /
• pp.415-422
• /
• 2024

Recent advances in construction automation have led to increased use of deep learning-based computer vision technology for construction monitoring. However, monitoring systems based on supervised learning struggle with recognizing complex risk factors in construction environments, highlighting the need for adaptable solutions. Large multimodal models, pretrained on extensive image-text datasets, present a promising solution with their capability to recognize diverse objects and extract semantic information. This paper proposes a methodology that generates training data for multimodal models, including safety-centric descriptions using GPT-4V, and fine-tunes the LLaVA model using the LoRA method. Experimental results from seven construction site hazard scenarios show that the fine-tuned model accurately assesses safety status in images. These findings underscore the proposed approach's effectiveness in enhancing construction site safety monitoring and illustrate the potential of large multimodal models to tackle domain-specific challenges.

• Proceedings of the IEEK Conference
• /
• 2002.06d
• /
• pp.373-376
• /
• 2002

This paper represents a survey of a variety thresholding techniques including both global and local thresholding. Several thresholding methods are examined in detail to evaluate their performance based on a given set of test images. We also attempt to evaluate the performance of several thresholding methods for construction field documents image recognition system using a broken line structures, broken symbols and text, blurring of lines, symbols and text, noise in homogeneous areas measure as a criterion functions.