• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2001년도 추계학술대회 학술발표 논문집
• /
• pp.51-55
• /
• 2001

With the rapid growth of the World Wide Web and electronic information services, information is becoming available on-Line at an incredible rate. One result is the oft-decried information overload. No one has time to read everything, yet we often have to make critical decisions based on what we are able to assimilate. The technology of automatic text summarization is becoming indispensable for dealing with this problem. Text summarization is the process of distilling the most important information from a source to produce an abridged version for a particular user or task. Information retrieval(IR) is the task of searching a set of documents for some query-relevant documents. On the other hand, text summarization is considered to be the task of searching a document, a set of sentences, for some topic-relevant sentences. In this paper, we show that document information, that is more reliable and suitable for query, using document length normalization of which is gained through information retrieval . Experimental results of this system in newspaper articles show that document length normalization method superior to other methods use query itself.

• 대한산업공학회지
• /
• 제29권3호
• /
• pp.197-205
• /
• 2003

Many people get plenty of information from World Wide Web, and the study of the factors that affect on reading task on web browser is presenting important issue. But domestic studies on legibility of Korean on web environment were relatively poor and the study about suitable text layout for skimming wasn't carried out also. At this point, this study was performed to investigate the effects of two types of reading, three levels of line length, and three levels of interlinear spacing on comprehension and reading rate when subjects read the materials on web browser. Reading speed, error rate, subjective preference and SACL(Stress and Arousal Checklist) evaluation were measured to evaluate the effects. Eighteen volunteer subjects participated in eighteen web document sessions with two different reading types, three different line lengths, and three different interlinear spacings. Statistical results from objective and subjective evaluations indicate that 50 characters per line of line length and 100 percents of interlinear spacing improved reading rate, overall error rates were reduced when reading normally, and SACL measures were increased at fast reading type. Consequently, in order to design text layout to retrieve information in WWW environment effectively, just applying guidelines of traditional printed material is not proper. Therefore, it is effective to consider reading type, line length, and interlinear spacing. Implications of these results and suggestions for the further study are also addressed.

• ETRI Journal
• /
• 제32권5호
• /
• 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.

• 정보처리학회논문지D
• /
• 제17D권6호
• /
• pp.463-470
• /
• 2010

본 논문에서는 DCT 데이터에서 영상 데이터로의 해독 및 이진화 과정을 생략하고 컬러 영상의 DCT 관련 원자료를 사용하는 방법에 기반을 둔 매우 빠르고 안정적인 문자열 구간 분리 모형을 제안하였다. DCT 블록에 저장된 DC 및 3개의 주요 AC 변수들을 조합하여 축소된 저해상도 회색 영상을 만들고 횡렬 및 종렬 투영법을 통해 얻어진 픽셀 값의 히스토그램을 분석하여 문자 열 구간 사이에 존재하는 백색의 띠 공간을 찾아내었다. 이 과정 중 탐색되지 않은 문자 열 구간은 마코프 모델을 사용하여 숨겨진 주기를 찾아내어 복원하였다. 본 논문에 실험 결과를 제시하였으며 기존의 방법보다 약 40 - 100배 빠른 방법임을 입증하였다.

• International Journal of Contents
• /
• 제9권2호
• /
• pp.8-13
• /
• 2013

Locating text data in images automatically has been a challenging task. In this approach, we build a three stage system for text detection purpose. This system utilizes tensor voting and Completed Local Binary Pattern (CLBP) to classify text and non-text regions. While tensor voting generates the text line information, which is very useful for localizing candidate text regions, the Nearest Neighbor classifier trained on discriminative features obtained by the CLBP-based operator is used to refine the results. The whole algorithm is implemented in MATLAB and applied to all images of ICDAR 2011 Robust Reading Competition data set. Experiments show the promising performance of this method.

• 정보화연구
• /
• 제11권1호
• /
• pp.63-73
• /
• 2014

웹의 발달로 텍스트 등으로 이루어진 비정형 데이터의 활용에 대한 관심이 높아지고 있다. 웹상에서 사용자들이 작성한 대부분의 비정형 데이터는 사용자의 주관이 담겨져 있어 이를 적절히 분석할 경우 사용자의 취향이나 주관적인 관점 등의 아주 유용한 정보를 얻을 수 있다. 이 논문에서는 이러한 비정형 텍스트 문서를 다양한 차원으로 분석하기 하는데 OLAP(온라인 분석 처리)의 다차원 데이터 큐브 기술을 활용한다. 다차원 데이터 큐브는 간단한 문자나 숫자 형태의 정형적인 데이터에 대해 다차원 분석하는데 널리 사용되었지만, 텍스트 문장으로 이루어진 비정형 데이터에 대해서는 활용되지 않았다. 이러한 텍스트 데이터베이스에 포함된 정보를 다차원으로 분석하기 위한 방법으로 텍스트 큐브 모델이 최근에 제안되었는데, 이 텍스트 큐브는 정보 검색에서 널리 사용하는 용어 빈도수(Term Frequency)와 역 인덱스(Inverted Index)를 측정값으로 이용하여 텍스트 데이터베이스에 대한 다차원 분석을 지원한다. 이 논문에서는 이러한 다차원 텍스트 큐브를 활용하여 실제 서비스되고 있는 호텔 정보 공유 사이트의 리뷰 데이터 분석에 활용하였다. 이를 위해 호텔 리뷰 데이터에 대한 다차원 텍스트 큐브를 생성하였으며, 이를 이용하여 다차원 키워드 검색 기능을 제공하여 사용자 중심의 의미있는 정보 검색이 가능한 시스템을 설계 및 구현하였다. 또한, 본 논문에서 제안하는 시스템에 대해 다양한 실험을 수행하였으며 이를 통해 제안된 시스템의 실효성을 검증하였다.

• 한국생물정보학회:학술대회논문집
• /
• 한국생물정보시스템생물학회 2000년도 International Symposium on Bioinformatics
• /
• pp.71-73
• /
• 2000

Considering that there is the lack of standards for storing genome-related on-line documents, the techniques in Natural Language Processing (NLP) is likely to become more and more important. It is necessary to extract useful information from the raw text and to store it in a computer-readable database format. Recent advances in NLP technologies raise new challenges and opportunities for tackling genome-related on-line text for information extraction task, For example, we can obtain many useful information related to genetic network or metabolic pathways simply by analyzing verbs such as 'activate'or 'inhibit'in Medline abstracts in a fully automatic way, Thus, combining NLP techniques with genome informatics extends beyond the traditional realms of either technology to a variety of emerging applications.

• 스마트미디어저널
• /
• 제6권4호
• /
• pp.32-40
• /
• 2017

In this paper, a robust text detection method based on edge enhanced contrasting extremal region (CER) is proposed using stroke width transform (SWT) and tensor voting. First, the edge enhanced CER extracts a number of covariant regions, which is a stable connected component from input images. Next, SWT is created by the distance map, which is used to eliminate non-text regions. Then, these candidate text regions are verified based on tensor voting, which uses the input center point in the previous step to compute curve salience values. Finally, the connected component grouping is applied to a cluster closed to characters. The proposed method is evaluated with the ICDAR2003 and ICDAR2013 text detection competition datasets and the experiment results show high accuracy compared to previous methods.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 ITC-CSCC -2
• /
• pp.835-838
• /
• 2000

Popular methods for extracting a text region in video images are in general based on analysis of a whole image such as merge and split method, and comparison of two frames. Thus, they take long computing time due to the use of a whole image. Therefore, this paper suggests the faster method of extracting a text region without processing a whole image. The proposed method uses line sampling methods, FFT and neural networks in order to extract texts in real time. In general, text areas are found in the higher frequency domain, thus, can be characterized using FFT The candidate text areas can be thus found by applying the higher frequency characteristics to neural network. Therefore, the final text area is extracted by verifying the candidate areas. Experimental results show a perfect candidate extraction rate and about 92% text extraction rate. The strength of the proposed algorithm is its simplicity, real-time processing by not processing the entire image, and fast skipping of the images that do not contain a text.

• 정보처리학회논문지D
• /
• 제17D권1호
• /
• pp.59-66
• /
• 2010

기존의 OCR 엔진은 보정된 환경에서 읽혀진 서류 영상에 맞게 설계되어있다. 스마트 폰을 비롯한 검정 화면 거리가 보정되지 않은 기기에서 읽혀진 영상에서는 삼차원 원근 투시에 의한 찌그러짐 또는 곡면상에서의 찌그러짐 등이 핵심적인 문제점들로 여겨진다. 휴대용 단말기에서 읽혀진 영상들에서의 OCR 기능에 대한 요구가 증가일로에 있는 시점에서, 본 논문에서는 문제점들을 세 가지로 구분하고 - 회전에 무관한 문자 영역 추출, 폰트 등의 크기에 무관한 문자 선 영역 추출, 3차원 매핑 이론 - 이를 해결하기위한 방법을 제시하였다. 이러한 방법론을 통합하여 카메라 영상 위에서의 OCR을 개발하였다.