• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.7 no.6
• /
• pp.1167-1174
• /
• 2006

Text region detection from a natural scene is useful in many applications such as vehicle license plate recognition. Therefore, in this paper, we propose a text region extraction method using pattern histogram of character-edge maps. We create 16 kinds of edge maps from the extracted edges and then, we create the 8 kinds of edge maps which compound 16 kinds of edge maps, and have a character feature. We extract a candidate of text regions using the 8 kinds of character-edge maps. The verification about candidate of text region used pattern histogram of character-edge maps and structural features of text region. Experimental results show that the proposed method extracts a text regions composed of complex background, various font sizes and font colors effectively.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.11B
• /
• pp.1020-1029
• /
• 2008

This paper proposes an high speed pattern matching algorithm and its implementation. The pattern matcher is used to check patterns from realtime input packet. The proposed algorithm can find exact string, range of string values, and combination of string values from input packet at high speed. Given string and rule set are modelled as a state transition graph which can find overlapped strings simultaneously, and the state transition graph is partitioned according to input implicants to reduce implementation complexity. The pattern matcher scheme uses the transformed state transition graph and input packet as an input. The pattern matcher was modelled and implemented in VHDL language. Experimental results show the proprieties of the proposed approach.

• Journal of the Korea Convergence Society
• /
• v.8 no.8
• /
• pp.33-40
• /
• 2017

In this paper, we propose a text extraction method by pattern analysis on contour for effective text detection in image. Text extraction algorithms using edge based methods show good performance in images with simple backgrounds, The images of complex background has a poor performance shortcomings. The proposed method simplifies the color of the image by using K-means clustering in the preprocessing process to detect the character region in the image. Enhance the boundaries of the object through the High pass filter to improve the inaccuracy of the boundary of the object in the color simplification process. Then, by using the difference between the expansion and erosion of the morphology technique, the edges of the object is detected, and the character candidate region is discriminated by analyzing the pattern of the contour portion of the acquired region to remove the unnecessary region (picture, background). As a final result, we have shown that the characters included in the candidate character region are extracted by removing unnecessary regions.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2001.12a
• /
• pp.295-298
• /
• 2001

본 논문에서는 CCD카메라로 입력 받은 다중 크기 및 활자체로 구성된 한글문서의 화상 데이터를 편집기에서 수정 가능한 문자로 변환시키는 시스템을 구현하였다. 먼저 Dynamic 이 진화 처리 과정을 거친 화상을 흑백 화소의 누적분포에 따라 문자단위로 분할한 후, 다양한 크기로 분할된 문자를 표준패턴 크기로 표준화 시켰다. 한글을 자소 간 공백 위치의 특징에 따라서 6가지 유형으로 분류한 후, 퍼지 이론을 접목시킨 원형 패턴 벡터 알고리즘을 사용해서 표준벡터와 입력된 글자의 특징벡터를 비교하여 문자로 인식하게 하였다. 각 6가지 유형에서 서로 다른 자소로 결합된 문자들을 30개 선정하여 여러 가지 활자체 및 크기에 적용해 본 결과, 모두 문서화가 가능함을 알 수 있었다.

• Journal of the Korea Computer Industry Society
• /
• v.2 no.3
• /
• pp.285-296
• /
• 2001

This Paper develops the Hanguel character learning system for elementary school students and foreigners. Standard character pattern is selected and DB is consructed for model by feature extraction. For this, performance of pre-processing independent of environments, feature extraction and simility functions are defined. Finally, beauty evaluation is done by matching between input character pattern written by elementary school students or foreigners and standard character pattern. It is possible for this system to extend the specific character font learning from selecting the specific standard character pattern. Also the effectiveness of this parer is demonstrated by several experiments.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06b
• /
• pp.327-329
• /
• 2012

문서 내에 존재하는 중요한 개체들 간의 관계를 자동으로 추출할 때 개체와 개체 사이의 상호작용 표현에 중요하게 관여하는 핵심자질을 잘 선택할수록 빠르고 정확하게 관계 추출을 수행할 수 있다. 본 논문에서는 개체 쌍 사이에 존재하는 술어-논항 구조의 어휘 패턴 문자열을 정규화해서 스트링 커널에 적용하는 관계 추출 방법을 제안한다. 제안된 시스템의 성능 평가를 위해서 과학기술문헌에 존재하는 중요한 개체들 간의 연관관계 추출 성능 평가를 수행하는 테스트컬렉션을 자체적으로 구축하였으며 실험을 통하여 제안된 방법의 성능을 측정하였다. 정확도 실험 결과, 스트링 커널의 입력으로 문장 전체를 사용한 경우에는 55.0693%, 개체 쌍 사이의 문자열을 사용한 경우에는 61.0331%, 그리고 술어-논항 구조의 어휘 패턴 문자열을 사용한 경우에는 69.14%로, 술어-논항 구조의 어휘 패턴 문자열을 사용했을 때 성능이 가장 좋게 나타났다. 결론적으로 문장 내의 술어-논항 구조를 분석하여 정규화된 어휘 패턴을 생성하고 이렇게 생성된 문자열을 스트링 커널에 적용하는 방법이 관계 추출에 유용한 방법임을 알 수 있었다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.32 no.9
• /
• pp.475-482
• /
• 2005

String data containing wildcard characters may represent certain patterns in texts. A subsumption relation between two patterns can be defined by a subset relation between sets of strings that match those patterns. Thus, the subsumption relation check is important to determine whether each pattern represents a set of strings without any overlap with another pattern. In this paper, we propose an effective algorithm that can determine subsumption relation between strings with wildcard characters. First, we consider a simple extension of the suffix tree algorithm so that it nay include wildcard characters and then we propose another method that checks the subsumption relation by dividing a suffix tree structure at each location of string data.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2003.05b
• /
• pp.635-640
• /
• 2003

본 논문에서는 일본 차량 번호판 인식에 적응적 탬플릿 마스킹 방법을 이용하여 번호판 문자, 숫자를 분할하고 패턴벡터기법을 이용하여 인식하는 방법을 제안하였다 주, 야간과 거리에 따른 일본 차량 번호판 영상을 입력받아 전처리 과정을 수행한 후 에지 정보와 명도값 변화의 빈도수를 이용하여 번호판 영역을 검출하였다 검출된 번호판 영역에서 각 문자 및 숫자의 위치정보와 적응적 탬플릿을 이용하여 분할하고 번호판의 지역문자를 무게중심 패턴으로 분류 한 다음 크기와 이동에 무관한 특실을 가지는 패턴 벡터를 적용하여 문자를 인식하였으며, 숫자는 Four Segment Pattern을 이용하여 인식하도록 하였다 본 논문에서 제안한 방법을 실제 일관 차량 번호판 인식에 적용한 결과 98.8％ 추출율과 96.6％의 인식율을 나타내었다.

• The Journal of the Korea Contents Association
• /
• v.2 no.1
• /
• pp.59-67
• /
• 2002

This study pre-processed the characters, performed the feature extraction for the beauty evaluation, and then defined the similarity function. It suggested the definition of the similarity function, and the extraction of the features of character elements. it experimented how much the various input character patterns were similar with the standard character patterns, found their results were almost similar with the expected ones and the results of beauty evaluation on general people through the questionaire with the results of the methods suggested here.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.12 no.4
• /
• pp.306-310
• /
• 2002

In this paper, we realized a system that converts the Korean alphabet (Hangul) images, which are in any domain that is formed by the finger movement on the Hangul document, to the editable characters and then outputs them to the word editor. The domain of hand is separated from the sphere of document in the pre-process step of image. The centroid point of hand is drawn by the maximum circular movement method. After the system recognizes the hand with the circular pattern vector algorithm, finds out the position of finger by the distance spectrum and then draws out the sphere of selected character image by the finger movement to divide the characters into character units by applying the histogram between the Hangul characters. We standardized the characters of various sizes. We used the circular pattern vector algorithm that grafts on the fuzzy inference to divert the character images of the domain, which user wants, to the editable characters by comparing the characteristic vectors between the standard pattern character and the inputted character and by recognizing the character.