• Title/Summary/Keyword: 채색

Search Result 312, Processing Time 0.023 seconds

A Study on Colorization of Buddhist Art Using Deep Learning (딥러닝을 이용한 불교 미술 채색 연구)

  • Cha, Jin-Hyuck;Jung, Seung-Won
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2018.05a
    • /
    • pp.392-393
    • /
    • 2018
  • 컬러 사진을 흑백으로 바꿔 데이터 셋을 구축해 CNN 모델을 트레이닝 시킨다. HSI 컬러 모델에서 흑백 사진으로부터 밝기 값은 알고 있음으로, 나머지 두 컬러 영역 색상, 채도를 CNN 모델이 추측하도록 학습을 시켜, 흑백사진으로부터 컬러 사진을 만들어낸다. 이 기술을 채색이 안된 불교 미술 그림에 적용하여 채색을 시키는 것에 대한 연구를 진행하였다.

Image Colorization based on Optimization (최적화 기법에 기반한 정지 영상의 채색화 기법)

  • Heu, Jun-Hee;Hyun, Dye-Young;Lee, Sang-Uk
    • Proceedings of the Korean Society of Broadcast Engineers Conference
    • /
    • 2008.11a
    • /
    • pp.207-210
    • /
    • 2008
  • 채색화는 흑백 영상에 색 정보를 추가하거나 영상의 색을 변환하는 영상 개선 기법이다. 본 연구는 최소한의 사용자 개입을 통해 흑백 영상을 자연스러운 칼라 (color) 영상으로 전환하는 채색화 기법을 제안한다. 우리는 우선 자연스러운 채색 결과를 위한 채색화 함수를 정의한다. 제안하는 채색화 함수는 유사한 밝기 정보를 가지는 이웃 픽셀들은 비슷한 색 정보를 가질 확률이 높다는 간단한 가정 하에 MRF (Markov Random Field)에 기반하여 모델링한다. 채색화 함수에 의해 색이 전체적으로 자연스럽게 분포될 수 있도록, 확산 신뢰도를 정의한 후 신뢰도에 따라 채색 순서를 결정한다. 이후, 채색 순서에 따라 각 픽셀에 채색화 함수를 적용하여 자연스러운 채색 결과를 도출한다. 실험 결과에서 보듯이, 제안 기법은 적은 색상 정보의 입력을 통해 효과적으로 채색화 하며, 기존 기법에 비해 자연스러운 결과를 제시한다.

  • PDF

Color Transfer in Painting Images Using Separation of Saturation (채색 분리를 이용한 그림 영상에서의 색변환 기법)

  • Kwak, Jeong-Min;Park, Chan-Woo;Moon, Young-Shik
    • Proceedings of the Korean Information Science Society Conference
    • /
    • 2007.10c
    • /
    • pp.505-509
    • /
    • 2007
  • 색 변환(Color Transfer) 기법은 컴퓨터 비전 및 영상 처리 분야에서 점점 더 많은 연구가 되고 있는 분야이다. 이는 참조 영상의 분위기를 원본 영상에 반영하는 기법이다. 본 논문에서는 채도가 낮은 색상에서 나타나는 잘못된 연산 결과를 해결하기 위해 채도 문턱치에 따라 유채색과 무채색으로 분류하여 인덱싱 하고, Lab색 모델에서 색상 채널인 a, b를 사용하여 그림 영상에서의 색 변환하는 방법을 제안한다. 제안된 방법은 영상의 화소들의 채도 문턱치를 이용하여 유채색과 무채색으로 분류하는 단계, 분류된 화소들의 색 특성을 이용한 cylindrical metric를 이용한 인덱싱 하는 단계, 각 인덱스 내의 위치적 표준편차와 화소수를 이용하여 인덱스들의 우열을 가리는 단계, 인덱스들의 우세한 순서로 Lab 색 모델에서 a 채널과 b 채널을 이용하여 색 변환하는 단계 등 4단계로 구성된다. 실험결과는 제안한 방법이 무채색과 유채색이 잘 분류되어 인덱싱 되었음을 보이고 원본 영상의 색이 참조영상의 색으로 잘 변환된 것을 보인다.

  • PDF

Comparison Analysis on Automatic Coloring System Algorithm Using Machine Learning (머신러닝을 활용한 자동 채색 시스템 알고리즘 비교 분석)

  • Lee, song eun;Lee, Ji Yeon;Kim, Na Heon;Kim, Jin Hwan
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2017.11a
    • /
    • pp.792-794
    • /
    • 2017
  • 현재 머신러닝(Machine Learning) 기술은 기존의 머신러닝과 조합 및 변형 되어 조금 더 발전 된 형태로 연구되어지고 있다. 따라서 수많은 알고리즘이 개발되고 있는 시점이다. 본 연구는 최근 좋은 결과로 관심을 받고있는 GAN(Generative Adversarial Net)을 중심으로 IT기술의 머신러닝과 그림을 조합하여 자동채색을 목적으로 GAN 알고리즘을 비교하고 분석하고자 한다. GAN 알고리즘들 가운데서 'Conditional GAN'과 'Wasserstein GAN'을 사용하여 자동채색을 적용시켰고, 가장 부합한 알고리즘을 찾고 성능을 비교하여 어떠한 알고리즘이 '자동채색' 목적에 더 부합한지 비교하고 판단 한다.

An Efficient Hint Input Method for Automatic Color Painting (선화 자동채색을 위한 효율적인 힌트입력 기법)

  • Kim, Taehoon;Lee, Seongjin
    • Proceedings of the Korean Society of Computer Information Conference
    • /
    • 2020.01a
    • /
    • pp.21-22
    • /
    • 2020
  • 본 논문에서는 자동채색 머신러닝 모델을 모바일에서 사용하기 위해 효율적으로 힌트 입력하기 위한 방법을 제시한다. 기존의 자동채색 기법은 서버-클라이언트 채색 방식으로 인터넷에 연결되어 있어야만 채식이 된다는 문제가 있다. 이를 보완하여 네트워크 연결 없이 모바일에서 사용할 수 있는 머신러닝과 이를 사용하기 위해 입력으로 넣어야 하는 힌트 및 이미지를 효율적으로 나눌 수 있는 방법을 고안한다.

  • PDF

Study on Coloring and Aging Pattern of Bongchae used for Conservation Treatment of Paper Cultural Properties (지류문화재의 보채에 쓰이는 봉채의 채색 및 열화 양상 연구)

  • Song, Jung Won;Kim, Myoung Nam;Lee, Jang Jon
    • Journal of Conservation Science
    • /
    • v.35 no.5
    • /
    • pp.416-429
    • /
    • 2019
  • Bongchae is primarily employed for toning, which is the last step in the conservation treatment of paper's cultural properties. The objective of this study is to identify the coloring patterns of four types of Bongchae(Bonram, Gamboge, Yeonji, and Daeja) and determine the associated color change patterns through accelerated aging experiments. By examining the coloring patterns, it is observed that Yeonji, Bonram, and Daeja are painted as particles, whereas Gamboge indicates a close state of coating. Results obtained from X-ray diffraction analysis of Bongchae indicate that the presence of quartz, calcite, and pyrophyllite can be equally detected in Bonram, Yeonji, and Daeja. Additionally, the presence of goethite is also detected in Daeja. Gamboge becomes discolored from yellow to orange color during wet thermal aging, and Gamboge and Yeonji become decolorized during UV irradiation aging. Hence, cultural properties of paper colored with Gamboge can be predicted to become discolored to orange color by alkalis and can be darkened by UV rays because the decolorization of Gamboge and Yeonji occurs preferentially.

Estimation of Spectral Distribution of Illumination Using Maximum Achromatic Region (최대 무채색 영역을 이용한 광원의 분광분포 추정)

  • Kim, Hui-Su;Kim, Yun-Tae;Lee, Cheol-Hui;Ha, Yeong-Ho
    • Journal of the Institute of Electronics Engineers of Korea SP
    • /
    • v.38 no.4
    • /
    • pp.392-400
    • /
    • 2001
  • This paper proposes an illuminant estimation algorithm that estimates the spectral power distribution of an incident light source from a single image. The proposed illumination recovery procedure has two phases. First, the surface spectral reflectances are recovered in the maximum achromatic region (MAR) which is the most achromatic and highly bright region of an image after removing partially the effect of illumination using a modified gray world algorithm. Here, the surface reflectances of MAR are estimated using the principal component analysis method along with a set of given 1269 Munsell samples. Second, the Population of reflected lights is determined with 1269 Munsell samples and a set of illuminations then the spectral distribution of re(looted lights of MAR is selected from the spectral database. That is, color differences are compared between the reflected lights of the MAR and the spectral database, which is the set of reflected lights built by the given set of Munsell samples and illuminants. Then the closest colors from the spectral database are selected. Finally, the illuminant of an image can be calculated dividing the average spectral distributions of reflected lights of MAR by the average surface reflectances of the MAR. In order to evaluate the proposed algorithm, experiments with artificial scenes, which are exposed to chromatic illuminants, were performed and the spectral distribution of estimated illumination and color difference are compared with results of the conventional method.

  • PDF

Mineralogical Characteristics and Provenance of Cu-bearing Green Minerals Used as Traditional Pigments (전통 안료로 사용된 구리함유 녹색광물의 광물학적 특성과 산지추정)

  • Do, Jin-Young;Jung, Jongmee
    • Journal of the Mineralogical Society of Korea
    • /
    • v.31 no.2
    • /
    • pp.123-135
    • /
    • 2018
  • This study tried to find the clue to Seokrok province by comparing Seokrok used in painting culture properties with Seokrok ore from domestic occurrence and imported Seokrok ore. To this end, chemical and mineralogical characteristics of painting cultural properties were identified with portable X-ray Florescence (p-XRF), micro X-ray diffraction (micro XRD) and SEM/EDS Analysis. To obtain Pb isotopic ratio, the Pb contained in Seokrok has been analyzed with Thermal Ionization Mass Spectrometer. Atacamite (or botallackite) and small quantity of brochantite were identified from Seokrok in Dancheong, and malachite was also identified from Buddhist painting besides those two ingredients. Without distinction of type, most Seokrok used in painting cultural properties is atacamite composed of Cu and Cl. From Pb isotope analysis, it was found that Seokrok in painting cultural properties was closer to that of north Korea, north China and Japan than south Korea as in regional division for East North Asia suggested by Mabuchi. The Pb isotopic ratio of domestic green mineral belongs to the distribution of Seokrok inside the painting cultural properties but imported malachite showed considerably difference. Considering the fact that atacamite, the main mineral of Seokrok in painting cultural properties is rarely produced from southern mine of the Korean Peninsula and the result of Pb isotope analysis.

Study of Lime Finishing Layer(intonaco) in Koguryo Tomb Murals (고구려 고분벽화의 석회마감층에 관한 연구)

  • Lee, Hwa-Soo;Han, Kyeong-Soon
    • Journal of Conservation Science
    • /
    • v.19
    • /
    • pp.43-56
    • /
    • 2006
  • We checked the conditions and material properties after producing the lime finishing layers with a mortar mix of processed lime. In addition, we reproduced the lime finishing layers of the Koguryo Tomb Murals, colored the layers with a few different techniques and identified the characteristics for each painting technique. As a result of the experiment, we could infer the producing techniques as well as teaming the conditions and properties of the lime finishing layers of the Koguryo tomb murals, and examined the maintenance of the lime finishing layers. Of those samples produced under various conditions, the layers of slaked lime, which were obtained after being hydrated underground for a long time, showed quite a good condition. Thus the slaked lime mixes produced through particular hydration methods are judged to increase the maintenance of the lime finishing layers of the Koguryo Tomb Murals. Mostly, the paints of Koguryo Tomb Murals indicated that they have similar conditions with the samples produced with a secco method. In addition, Considering the compositional forms of the painting layers and the boundary layers which formed between the painting layers and the finishing layers, we can verify that, in some cases, the conditions of the painting layers of the Koguryo tomb murals are similar to those which were produced with a combination of fresco and secco.

  • PDF

Application of Handheld Raman Spectroscopy for Pigment Identification of a Hanging Painting at Janggoksa Temple(Maitreya Buddha) (장곡사 미륵불 괘불탱의 채색 재료 분석을 위한 휴대용 라만 분광기의 적용성 연구)

  • LEE Na Ra;YOO Youngmi;KIM Sojin
    • Korean Journal of Heritage: History & Science
    • /
    • v.56 no.4
    • /
    • pp.216-228
    • /
    • 2023
  • The purpose of this study is to apply the handheld Raman spectrometer to identify the coloring materials used in a large Buddhist painting (of Maitreya Buddha) at Janggoksa Temple through cross-validation with HH-XRF. An in situ investigation was performed together with use of a digital microscope and HH-XRF analysis to verify the properties of pigments used in the gwaebul ("large Buddhist painting") via a non-destructive method. However, the identification of coloring materials composed of light elements and mixed or overlaid pigments is difficult using only non-destructive analysis data. Unlike in situ investigation, laboratory analysis often required samples yet the sampling is restricted to a small quantity due to the cultural heritage characteristic. Thus, it is necessary to develop a non-destructive in situ method to supplement the HH-XRF data. The large Buddhist painting at Janggoksa Temple was painted mainly using white, red, yellow, green, and blue colors. The Raman spectroscopy provides molecular information, while XRF spectroscopy provides information about elemental composition of the pigments. Analysis results identified various coloring materials: inorganic pigment, such as lead white, minium, cinnabar, and orpiment, as well as organic pigment such as gamboge and indigo. Therefore, it is possible to obtain more information for the identification of pigments; organic pigment and mixed or overlaid pigments, while at the same time minimizing the collection sample and simplifying the analysis procedure compared to previously used methods. The results of this study will be used as basic data for the analysis of painting cultural heritage through a non-destructive in situ method in the future.