• Title/Summary/Keyword: 획 복원

Search Result 9, Processing Time 0.024 seconds

Restoration of Numeral Strings Touched with Lines in Various Form Documents (서식 문서의 선과 접촉된 숫자열 복원에 관한 연구)

  • Lee, Chang-Hyeon;Choe, Yeong-U;Kim, Gyeong-Hwan;Lee, Il-Byeong
    • Journal of KIISE:Software and Applications
    • /
    • v.28 no.6
    • /
    • pp.439-449
    • /
    • 2001
  • 본 논문에서는 서식 문서의 선과 숫자의 획이 접촉된 경우 숫자의 획을 접촉되기 전 상태의 원 이미지로 복원하는 방법을 제안한다. 제안하는 방법에는 서기 문서에서 추출한 숫자열을 대상으로 열 단위로 복원한다. 과정은 우선 숫자열과 접촉된 선의 위치를 찾아내고, 선을 추적하면서 접촉으로 판정되는 영역을 유형별로 분류하여, 각 유형에 적합한 획 복원 방법을 제안한다. 또한 선에 숫자의 획이 완전히 포함된 경우의 복원 방법도 제안하여 현장에서의 서식 처리 과정에서 발생하는 문제점을 해결하고자 하였다. 제안하는 방법을 평가하기 위해서 은행 입출금전표, 신용카드 매출전표 및 NIST 필기 숫자열 데이터베이스 이미지를 사용하였다.

  • PDF

Automatic Stroke Extraction of TrueType Font and Handwriting of Hangul (한글 트루타입폰트 및 손글씨의 자동 획 분할 알고리즘)

  • Kwak, Yoon-Seok;Koo, Sang-Ok;Jung, Soon-Ki
    • Proceedings of the Korean Information Science Society Conference
    • /
    • 2008.06b
    • /
    • pp.275-280
    • /
    • 2008
  • 본 논문에서는 한글 글립(glyph)의 형태학적 분석을 통해 자동으로 획을 분할하는 방법을 제안한다. 제안된 방법은 thinning된 한글 글립의 골격(skeleton) 이미지를 기반으로, 획 분리, 획 병합, 그리고 획 볼륨 복원의 세가지 단계를 거쳐 한글의 기본 획들을 추출해 낸다. 실험 결과, 트루타입폰트(TrueType Font)에 대해서는 80%, 손글씨(Handwriting) 글립에 대해서는 72%의 획 분할 정확도를 보였다. 본 논문에서 제안한 방법으로 획득된 획 정보를 이용하여, 향후 한글 손글씨 생성을 위한 연구를 하고자 한다.

  • PDF

Surface profile measurement with optically trapped micro-particles (광포획된 마이크로 입자를 이용한 표면형상 측정)

  • 주지영;김준식;김승우
    • Proceedings of the Optical Society of Korea Conference
    • /
    • 2001.02a
    • /
    • pp.116-117
    • /
    • 2001
  • 정밀 삼차원 미세 형상 측정기에서 성능의 관건은 고속, 고분해능으로 측정하는 것이다. 그러기 위해서는 공진주파수가 높아야 하고 스프링 상수가 작아야 한다. 광포획 현미경(optical trap microscope, OTM)은 광포획 된 마이크로 입자를 프로브로 사용하는 것으로 입자에 작용하는 복원력이 광에 의한 힘뿐이므로 스프링 상수가 낮다. 또한 공진주파수는 f=√k/m 으로 입자의 질량이 매우 작으므로 공진주파수도 비교적 높다. (중략)

  • PDF

Recovery of Erased Character Strokes in the Extraction of Text Using Color Information (칼라정보에 기반한 텍스트 영역 추출에서의 지워진 획 복구)

  • Kim Seon-Hyung;Kim Ji-Soo;Kim Soo-Hyung
    • Annual Conference of KIPS
    • /
    • 2006.05a
    • /
    • pp.657-660
    • /
    • 2006
  • 자연영상이나 스팸메일 영상으로부터 텍스트 영역을 추출하고 추출한 텍스트 영역에 이진화를 수행하고 나면 가로 방향이나 세로획 방향으로 놓여 있는 "1" 그리고 "ㅡ" 에 해당하는 한글의 종성부분이 이미지 내의 잡영을 지울 때 종종 지워지는 결과를 볼 수 있다. 이렇게 지워진 획 부분을 되살리기 위한 방법으로 텍스트 Hinting 알고리즘을 제안한다. 텍스트 Hinting 알고리즘은 이진화된 이미지의 텍스트 픽셀 위치와 동일한 좌표에 해당하는 원본 이미지의 RGB 값을 추출하여 추출된 텍스트 후보 영역의 색상을 알아낸다. 추출된 텍스트 색상 레이어 이미지와 이진화된 이미지에 OR연산을 수행하게 되면 지워진 획 부분을 복원할 수 있다. 제안한 방법을 스팸 이미지에 적용한 결과 텍스트 추출결과를 획기적으로 개선할 수 있음을 보였다.

  • PDF

Variance Recovery in Text Detection using Color Variance Feature (색 분산 특징을 이용한 텍스트 추출에서의 손실된 분산 복원)

  • Choi, Yeong-Woo;Cho, Eun-Sook
    • Journal of the Korea Society of Computer and Information
    • /
    • v.14 no.10
    • /
    • pp.73-82
    • /
    • 2009
  • This paper proposes a variance recovery method for character strokes that can be missed in applying the previously proposed color variance approach in text detection of natural scene images. The previous method has a shortcoming of missing the color variance due to the fixed length of horizontal and vertical windows of variance detection when the character strokes are thick or long. Thus, this paper proposes a variance recovery method by using geometric information of bounding boxes of connected components and heuristic knowledge. We have tested the proposed method using various kinds of document-style and natural scene images such as billboards, signboards, etc captured by digital cameras and mobile-phone cameras. And we showed the improved text detection accuracy even in the images of containing large characters.

3D On-line Handwriting Character Recognition System for Wearable Devices (웨어러블 장치를 위한 3D 온라인 필기인식 시스템)

  • Kim, Minji;Choi, Lynn
    • Annual Conference of KIPS
    • /
    • 2014.11a
    • /
    • pp.1100-1103
    • /
    • 2014
  • 본 논문에서는 웨어러블 장치에서 펜 형태 또는 손가락 부착 형태의 입력 인터페이스로 사용할 수 있는 3D 온라인 필기인식 시스템을 제안한다. 3 축 가속도 센서와 자이로 센서를 장착한 입력 인터페이스를 사용하여 사용자는 손의 움직임을 통해 웨어러블 기기 또는 스마트 기기에 문자를 입력할 수 있다. 본 연구에서 제안하는 3D 필기인식 시스템은 필기 경로를 복원하여 획을 추출하고, 3 차원 공간의 필기문자에서 나타나는 기울임이나 왜곡, 겹쳐 쓰기를 고려한 특징점 추출 과정을 거친다. 추출한 특징점을 2 단계 결정 트리의 입력으로 사용하여 사용자가 공간상에서 필기한 알파벳을 인식한다. 10 명의 사용자에게 3 회의 필기 데이터를 입력 받아 총 780 개의 문자를 인식한 결과, 87.69%의 인식률을 얻을 수 있었다.

A Flexible Line-Fitting ICM Approach for Takbon Image Restoration (유연한 선부합 ICM 방식에 의한 탁본영상복원)

  • Hwang, Jae-Ho
    • The KIPS Transactions:PartB
    • /
    • v.13B no.5 s.108
    • /
    • pp.525-532
    • /
    • 2006
  • This paper proposes a new class of image restoration on the Ising modeled binary 'Takbon' image by the flexible line-fitting ICM(Iterated conditional modes) method. Basically 'Takbon' image need be divided into two extreme regions, information and background one due to its stroke combinations. The main idea is the line process, comparing with the conventional ICM approaches which were based on partially rectangular structured point process. For calculating geometrical mechanism, we have defined line-fitting functions at each current pixel array which form the set of linear lines with gradients and lengths. By applying the Bayes' decision to this set, the region of the current pixel is decided as one of the binary levels. In this case, their statistical reiteration for distinct tracking between intra and extra region offers a criterion to decide the attachment at each step. Finally simulations using the binary 'Takbon' image are provided to demonstrate the effectiveness of our new algorithm

A Survey on the Cheolwon Castle of Taebong-guk During the Japanese Colonial Period (일제강점기 태봉국 철원성 조사와 봉선사지)

  • Sim, Jaeyoaun
    • Korean Journal of Heritage: History & Science
    • /
    • v.52 no.1
    • /
    • pp.258-271
    • /
    • 2019
  • Recent changes in inter-Korean relations have encouraged South Korean archaeologists' interest in northern cultural heritage and their desire to conduct research. However, it is doubtful how much the South Korea really knows about the cultural ruins in North Korea and the cultural ruins to be found in the DMZ. In Korea, research data on the Japanese colonial period is scattered amongst national institutions and it is not possible to to collect, identify and document the data available in each institution. Typically, the data of Keiichi Ogawa(小川敬吉) is difficult to understand in its printed from. In addition, Ogawa has collected data from several national government agencies, and work is underway to collect architectural data and construct digital archives. This situation will not occur if we publish the data collected so far in digital form. Therefore, there is an urgent need to fully disclose the data related to Cheolwon which is owned by national institutions. If this data is analyzed, sufficient archaeological information can be obtained without excavation. On the other hand, one must wonder if the whole landscape of Cheolwon castle of Taebongguk cand be understood by investigating the interior of Cheolwon castle. This effort should be preceded by a survey on the ruins and the remains of the southern part of the southern boundary line. Rapid development and large-scale arrangement of cultivated land are likely to make it difficult to restore the external landscape related to the tillage demand that is required by the current situation. In the process of restoration of the actual Gyeongwon Line, it has been confirmed that serious irregularity has caused damage and the landscape has been transformed. In order to minimize these risks, it is urgent that we investigate the relics scattered in the southern part of the country, and not devote ourselves to the investigation of Cheolwon. In this regard, how much military, roads, excavation and archaeologists are prepared, centered on Cheolwon and scenery is an important question. I am curious as to how much preparation is provided to the Chulwon-gun and Gangwon-do in terms of administrative assistance, and how much archaeologists and excavation agencies are aware of the archaeological information of the inside and the southern part of the DMZ. Furthermore, how long have people been aware of the archaeological remains on the North Korean side? In order for da iscussion on Cheolwon and scenery to progress, it is necessary to carry out a precise investigation and accumulation of data on the remains in South Korea.

Excavation of Kim Jeong-gi and Korean Archeology (창산 김정기의 유적조사와 한국고고학)

  • Lee, Ju-heun
    • Korean Journal of Heritage: History & Science
    • /
    • v.50 no.4
    • /
    • pp.4-19
    • /
    • 2017
  • Kim Jeong-gi (pen-name: Changsan, Mar. 31, 1930 - Aug. 26, 2015) made a major breakthrough in the history of cultural property excavation in Korea: In 1959, he began to develop an interest in cultural heritage after starting work as an employee of the National Museum of Korea. For about thirty years until he retired from the National Research Institute of Cultural Heritage in 1987, he devoted his life to the excavation of our country's historical relics and artifacts and compiled countless data about them. He continued striving to identify the unique value and meaning of our cultural heritage in universities and excavation organizations until he passed away in 2015. Changsan spearheaded all of Korea's monumental archeological excavations and research. He is widely known at home and abroad as a scholar of Korean archeology, particularly in the early years of its existence as an academic discipline. As such, he has had a considerable influence on the development of Korean archeology. Although his multiple activities and roles are meaningful in terms of the country's archaeological history, there are limits to his contributions nevertheless. The Deoksugung Palace period (1955-1972), when the National Museum of Korea was situated in Deoksugung Palace, is considered to be a time of great significance for Korean archeology, as relics with diverse characteristics were researched during this period. Changsan actively participated in archeological surveys of prehistoric shell mounds and dwellings, conducted surveys of historical relics, measured many historical sites, and took charge of photographing and drawing such relics. He put to good use all the excavation techniques that he had learned in Japan, while his countrywide archaeological surveys are highly regarded in terms of academic history as well. What particularly sets his perspectives apart in archaeological terms is the fact that he raised the possibility of underwater tombs in ancient times, and also coined the term "Haemi Culture" as part of a theory of local culture aimed at furthering understanding of Bronze Age cultures in Korea. His input was simply breathtaking. In 1969, the National Research Institute of Cultural Heritage (NRICH) was founded and Changsan was appointed as its head. Despite the many difficulties he faced in running the institute with limited financial and human resources, he gave everything he had to research and field studies of the brilliant cultural heritages that Korea has preserved for so long. Changsan succeeded in restoring Bulguksa Temple, and followed this up with the successful excavation of the Cheonmachong Tomb and the Hwangnamdaechong Tomb in Gyeongju. He then explored the Hwangnyongsa Temple site, Bunhwangsa Temple, and the Mireuksa Temple site in order to systematically evaluate the Buddhist culture and structures of the Three Kingdoms Period. We can safely say that the large excavation projects that he organized and carried out at that time not only laid the foundations for Korean archeology but also made significant contributions to studies in related fields. Above all, in terms of the developmental process of Korean archeology, the achievements he generated with his exceptional passion during the period are almost too numerous to mention, but they include his systematization of various excavation methods, cultivation of archaeologists, popularization of archeological excavations, formalization of survey records, and promotion of data disclosure. On the other hand, although this "Excavation King" devoted himself to excavations, kept precise records, and paid keen attention to every detail, he failed to overcome the limitations of his era in the process of defining the nature of cultural remains and interpreting historical sites and structures. Despite his many roles in Korean archeology, the fact that he left behind a controversy over the identity of the occupant of the Hwangnamdaechong Tomb remains a sore spot in his otherwise perfect reputation.