• 한국멀티미디어학회논문지
• /
• 제18권10호
• /
• pp.1197-1204
• /
• 2015

Smoke detection plays an important role for the early detection of fire. In this paper, we suggest a newly developed method that generated LBPV(Local Binary Pattern Variance)s as special feature vectors from RGB contrast images can be applied to detect smoke using SVM(Support Vector Machine). The proposed method rearranges mean value of the block from each R, G, B channel and its intensity of the mean value. Additionally, it generates RGB contrast image which indicates each RGB channel’s contrast via smoke’s achromatic color. Uniform LBPV, Rotation-Invariance LBPV, Rotation-Invariance Uniform LBPV are applied to RGB Contrast images so that it could generate feature vector from the form of LBP. It helps to distinguish between smoke and non smoke area through SVM. Experimental results show that true positive detection rate is similar but false positive detection rate has been improved, although the proposed method reduced numbers of feature vector in half comparing with the existing method with LBP and LBPV.

• International Journal of Naval Architecture and Ocean Engineering
• /
• 제6권4호
• /
• pp.840-866
• /
• 2014

Visibility in underwater images is usually poor because of the attenuation of light in the water that causes low contrast and color variation. In this paper, a new approach for underwater image quality improvement is presented. The proposed method aims to improve underwater image contrast, increase image details, and reduce noise by applying a new method of using contrast stretching to produce two different images with different contrasts. The proposed method integrates the modification of the image histogram in two main color models, RGB and HSV. The histograms of the color channel in the RGB color model are modified and remapped to follow the Rayleigh distribution within certain ranges. The image is then converted to the HSV color model, and the S and V components are modified within a certain limit. Qualitative and quantitative analyses indicate that the proposed method outperforms other state-of-the-art methods in terms of contrast, details, and noise reduction. The image color also shows much improvement.

• Current Optics and Photonics
• /
• 제6권3호
• /
• pp.304-312
• /
• 2022

Digital images of cultural relics captured using line scan cameras present limitations due to uneven intensity and low contrast. To address this issue, this report proposes a colour linear array image enhancement method that can maintain a constant colour. First, the colour linear array image is converted from the red-green-blue (RGB) colour space into the hue-saturation-intensity colour space, and the three components of hue, saturation, and intensity are separated. Subsequently, the hue and saturation components are held constant while the intensity component is processed using the established intensity compensation model to eliminate the uneven intensity of the image. On this basis, the contrast of the intensity component is enhanced using an improved local contrast enhancement method. Finally, the processed image is converted into the RGB colour space. The experimental results indicate that the proposed method can significantly improve the visual effect of colour linear array images. Moreover, the objective quality evaluation parameters are improved compared to those determined using existing methods.

• 한국콘텐츠학회논문지
• /
• 제13권1호
• /
• pp.9-18
• /
• 2013

본 연구는 감상자가 선호하는 화질의 파라미터를 정량화하기 위하여 선행 연구를 기반으로 한 물리적인 화질 평가 항목과 인지적 특성을 반영한 화질 평가 항목을 다이내믹 레인지, 컬러, 콘트라스트로 설정하였다. 그리고 이미지 감상자가 선호하는 콘텐츠별 화질의 재현 성능 범위를 구체화하였다. 그 결과 다이내믹 레인지를 의미하는 디지털 존시스템은 인물사진, 야경사진, 풍경사진이 6~10 stop의 범위를 나타냈다. 전체 RGB 평균은 인물사진(67.2~215.2), 야경사진(46~142), 풍경사진(52~185)으로 인물사진이 RGB 컬러 범위가 가장 넓게 나타났으며, 다음으로 풍경사진, 야경사진의 순서대로 나타났다. 전체 콘트라스트의 범위는 인물사진(196~589), 야경사진(131~575), 풍경사진(104~767)으로 나타났다. 특히 인물사진의 경우, 노출의 기준이 되는 피부톤이 ZONE V로 나타났지만 실제로 감상자들이 선호하는 피부톤의 밝기는 ZONE IV에 해당하였다. 또한 인물사진 전체 장면과 메인 피사체의 콘트라스트 비율이 1:1.2를 나타내어 감상자가 아웃포커스 효과를 선호한다고 판단할 수 있었다. 이와 같은 결과를 통하여 일반 감상자들이 선호하는 디지털 이미지의 화질 재현 성능 범위를 수치화시킬 수 있었다. 또한 디지털 카메라의 개발자에게 감상자가 선호하는 실제 다이내믹 레인지, 컬러, 콘트라스트의 구체적인 정보 범위를 제공하여 제품 개발에 반영될 수 있기를 기대한다.

• 로봇학회논문지
• /
• 제17권3호
• /
• pp.314-321
• /
• 2022

Depth information used in SLAM and visual odometry is essential in robotics. Depth information often obtained from sensors or learned by networks. While learning-based methods have gained popularity, they are mostly limited to RGB images. However, the limitation of RGB images occurs in visually derailed environments. Thermal cameras are in the spotlight as a way to solve these problems. Unlike RGB images, thermal images reliably perceive the environment regardless of the illumination variance but show lacking contrast and texture. This low contrast in the thermal image prohibits an algorithm from effectively learning the underlying scene details. To tackle these challenges, we propose multi-channel remapping for contrast. Our method allows a learning-based depth prediction model to have an accurate depth prediction even in low light conditions. We validate the feasibility and show that our multi-channel remapping method outperforms the existing methods both visually and quantitatively over our dataset.

• 한국농공학회:학술대회논문집
• /
• 한국농공학회 2003년도 학술발표논문집
• /
• pp.623-626
• /
• 2003

Image analysis was performed with two color systems, Red-Green-Blue (RGB) values and normalized Hue-Saturation-Intensity (HSI). We conducted field studies in Cheongju to determine canopy spectral reflectance and digital image analysis of rice. Spectral reflectance measurements made with a portable spectrometer(LI-1800) correlated with growing stage and digital images for rice. Images in which the color was specified by the common RGB coordinates could be used when there was a sharp contrast between the color of the rice and that of the field soil. In the absence of sharp contrast, identification of the rice covered area was much easer after the color had been transformed into HSI coordinates. This study introduced fundamental theories in digital image analysis and applied that for field situations rice.

• 대한기관식도과학회지
• /
• 제18권2호
• /
• pp.45-48
• /
• 2012

Background Neoplastic vessels tend to proliferate on the surface of malignant lesions in the aerodigestive tract. So, superficial malignant lesions can be detected earlier by enhancing mucosal vascular clarity. To enhance mucosal vascular clarity on endoscopic image, we developed an image processing algorithm of RGB (red-green-blue) channel substitution image (CSI). Methods Each pixel in original white light image (WLI) has its own value of red, green and blue channel. Various combinations of RGB channel substitution was tried on original WLI. Results To make superficial blood vessels darker than brighter background mucosa, in the CSI algorithm, RGB value in each pixel of WLI is substituted; red value to green one, green value to blue one. There was a good contrast between superficial mucosal vessels and background brighter mucosa in the CSI image. Conclusion By RGB CSI algorithm, WLI could be successfully converted to new images with enhanced mucosal vascular clarity. Using RGB CSI algorithm could provide added vascular visibility on original WLI.

• 한국정보통신학회논문지
• /
• 제9권3호
• /
• pp.616-621
• /
• 2005

본 논문에서는 변환영역 기반과 인간의 시각특성을 적용하여 압축에 강한 칼라 영상의 디지털 워터마킹 알고리즘을 제안하였다. 원영상을 RGB 채널로 분리한 후, HVS 특성을 고려하여 명암대비와 텍스처 특징을 분석한 후, 최적의 주파수영역을 선택하여 워터마크를 삽입하였다 전처리 과정은 2D DCT를 사용하였고, 워터마크는 시각적으로 인지가 가능한 특정 로고 형태의 이진 영상을 사용하였다. 외부공격 유형으로 JPEG 압축을 수행하여 실험한 결과, JPEG 압축 $60\%$까지 워터마크의 추출이 가능하였고 $90\%$ 이상의 상관도를 보였다.

• 한국멀티미디어학회논문지
• /
• 제12권7호
• /
• pp.947-952
• /
• 2009

본 논문에서는 DIP 연산자를 이용한 컬러 스케치 영상 생성 방법을 제안한다. 제안된 방법에서는 먼저 입력 RGB 컬러 영상을 HSV 컬러 영상으로 변환한 다음 밝기 성분인 V 성분 영상에 DIP 연산자를 적용하여 V 성분 스케치 영상을 추출한다. 추출된 V 성분 스케치 영상은 시각적 편의를 위하여 반전과 대비 신장 과정을 거친다. S 성분 영상은 출력 컬러 스케치 영상의 컬러가 입력 영상의 컬러와 같으면서 약간 진하게 나타나도록 향상 과정을 거친다. 이들 S 성분 및 V 성분 영상들은 원래의 H 성분 영상과 함께 RGB 컬러 영상으로 변환되어 출력 컬러 스케치 영상을 얻는다. 실험 결과 제안된 방법은 시험 영상에 대하여 원 영상의 컬러를 잘 살리면서 손으로 그린 스케치화와 유사한 컬러 스케치 영상을 생성함을 보여준다.

• 한국인쇄학회지
• /
• 제28권2호
• /
• pp.117-128
• /
• 2010

Advancements in digital image have put high quality digital camera into the hands of many image professionals and consumers alike. High quality digital camera images consist originally of raw which have a set of color rendering operation applied to them to produce good images. With color rendering, the raw file was converted to Adobe RGB and sRGB color space. Also color rendering can incorporate factor such as white balance, contrast, saturation. Therefore, in this paper we conduct a study on production of optimum profile considered color rendering in digital camera. To do the experiment, the images were Digital ColorChecker SG target and ColorChecker DC target. A profiling tool was ProfileMaker 5.03. The results were analyzed by comparing in color gamut of $CIEL^*a^*b^*$ color space and calculating ${\Delta}E^*_{ab}$. Also results were analyzed in terms of different $CIEL^*a^*b^*$ color space quadrants based on lightness, chroma.