• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.179-181
• /
• 2022

동적 비전 센서(Dynamic Vision Sensor)라고도 알려진 이벤트 카메라는 생체에서 영감을 받은 새로운 시각 센서이다. 고정된 속도로 이미지를 생성하는 기존 카메라와 달리 이벤트 기반 카메라의 픽셀은 독립적이고 비동기적으로 작동한다. 기존 프레임 기반 카메라보다 이벤트 기반 카메라가 움직임을 포착하는데 더 적합하며 모션 블러(Motion Blur)가 없고 시간 해상도가 높다는 이점을 통해 고속카메라로 활용할 수 있다. 본 논문은 이벤트 카메라의 높은 시간 해상도와 동적 범위, 낮은 지연시간, 전력 소비량의 이점을 활용하여 움직이는 물체를 모션 블러 없이 포착하는 이벤트 기반 물체 추적 시스템을 제안한다. 실험을 통해 전체 영상을 포착하는 기존 프레임 기반 카메라에 비해 밝기 변화에 따른 동적 변화만을 추적하는 이벤트 기반 카메라는 모션 블러가 없다는 점을 검증하였다.

• Journal of the Korea Computer Graphics Society
• /
• v.14 no.3
• /
• pp.1-9
• /
• 2008

This paper presents a novel optimization-based saliency-preserving method for converting color images to grayscale in a manner consistent with conventional approaches of black-and-white photographers. In black-and-white photography, a colored filter called a contrast filter has been commonly employed on a camera to lighten or darken selected colors. In addition, local exposure controls such as dodging and burning techniques are typically employed in the darkroom process to change the exposure of local areas within the print without affecting the overall exposure. Our method seeks a digital version of a conventional contrast filter to preserve visually-important image features. Furthermore, conventional burning and dodging techniques are addressed, together with image similarity weights, to give edge-aware local exposure control over the image space. Our method can be efficiently optimized on GPU. According to the experiments, CUDA implementation enables 1 megapixel color images to be converted to grayscale at interactive frames rates.

• Journal of Internet Computing and Services
• /
• v.18 no.1
• /
• pp.43-50
• /
• 2017

In this paper, we are proposing improved vanishing points detection and segments labeling methods for inside wall frame detection from indoor image of a piece of having a colour RGB. A lot of research related to recognizing the frame of artificial structures from the image is being performed due to increase in demand for AR technology. But detect the inside wall frame in indoor images have many objects that caused the occlusion is still a difficult issue. Inner wall frame detection methods are usually segment labeling methods and detect vanishing point methods are used together. In order to improve the vanishing point detection method we proposed using inner wall orthogonality which forms the cube. Also we proposed labeling method using tree based learning and superpixel based segmentation method for labelingthe segments in indoor images. Finally, in experiments have shown improved results about inside wall frame detection according to our methods.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.10
• /
• pp.110-116
• /
• 2017

This paper presents an effective demosaicking algorithm for color filter array (CFA) images acquired from single-sensor devices based on directional interpolation and nonlocal properties of the image. We interpolate the G channel considering diagonal directions as well as horizontal and vertical directions, using a small number of pixels to reflect local properties of the image. Then, we overcome image degradations, such as zipper effects near edges and false colors, by applying nonlocal means (NLM) filtering to the interpolated pixels. R and B channels are reproduced by using directional interpolation with information of the reconstructed G channel and NLM filtering. Experimental results for various McMaster images with high saturation and color changes show that the proposed algorithm accomplishes high PSNR compared with conventional methods. Moreover, the proposed method demonstrates better subjective quality compared with existing methods in terms of reduction of quality degradation, like false colors, and preservation of the image structures, such as edges and textures.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.6
• /
• pp.117-126
• /
• 2010

In the case of failure images with controlling lighting like backlighting and excessive frontlinghting, the compensation scheme for a specific area in an image is required. The interested region is first selected by user in our method to compensate the first frame. Then we define the matching function of brightness and energy function is proposed with weight of matching function and the relationship among the neighbors. Finally, the energy is minimized by the graph-cut algorithm to compensate the brightness of the first frame. Other frames are straightforwardly compensated using the results of the first frame. The brightness variations of the previous frame is transmitted to the next frame via motion vectors. The reliability of the brightness variation is calculated based on the motion vector reliability. Video compensation result is achieved by the process of the image case. Simulation show that the proposed algorithm provides more natural results than the conventional algorithms.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.185-191
• /
• 2022

In this study, we propose a novel super-resolution neural network based on U-Net, residual neural network, and sub-pixel convolution. To prevent the loss of detailed information due to the max pooling of U-Net, we propose down-sampling and connection using sub-pixel convolution. This uses all pixels in the filter, unlike the max pooling that creates a new feature map with only the max value in the filter. As a 2×2 size filter passes, it creates a feature map consisting only of pixels in the upper left, upper right, lower left, and lower right. This makes it half the size and quadruple the number of feature maps. And we propose two methods to reduce the computation. The first uses sub-pixel convolution, which has no computation, and has better performance, instead of up-convolution. The second uses a layer that adds two feature maps instead of the connection layer of the U-Net. Experiments with a banchmark dataset show better PSNR values on all scale and benchmark datasets except for set5 data on scale 2, and well represent local area patterns.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.7
• /
• pp.149-156
• /
• 2011

This paper proposes the effective edge selection algorithm for the rapid processing time and low memory usage of efficient graph-based image segmentation on mobile device. The graph-based image segmentation algorithm is to extract objects from a single image. The objects are consisting of graph edges, which are created by information of each image's pixel. The edge of graph is created by the difference of color intensity between the pixel and neighborhood pixels. The object regions are found by connecting the edges, based on color intensity and threshold value. Therefore, the number of edges decides on the processing time and amount of memory usage of graph-based image segmentation. Comparing to personal computer, the mobile device has many limitations such as processor speed and amount of memory. Additionally, the response time of application is an issue of mobile device programming. The image processing on mobile device should offer the reasonable response time, so that, the image segmentation processing on mobile should provide with the rapid processing time and low memory usage. In this paper, we demonstrate the performance of the effective edge selection algorithm, which effectively controls the edges of graph for the rapid processing time and low memory usage of graph-based image segmentation on mobile device.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2014.11a
• /
• pp.200-202
• /
• 2014

본 논문은 High bit-depth 를 위한 SIMD (Single Instruction, Multiple Data) 명령어 기반 보간 필터 고속화 방법을 제안한다. 픽셀 연산을 기반으로 하는 보간 필터링은 HEVC 복호화기에서 높은 복잡도를 차지하고 있지만 반복적인 산술연산을 수행하기 때문에 SIMD 를 이용한 고속화에 적합한 구조를 가지고 있다. 이러한 이유로 본 논문에서는 보간 필터 연산에 대하여 SIMD 명령어를 이용하여 메모리를 효율적으로 사용하여 고속화하는 방법을 제안한다. 제안하는 기술은 HEVC 참조 소프트웨어 HM 12.0-RExt 4.1 에 기반을 둔 ANSI C 기반 자체 개발 HEVC RExt 복호화기 소프트웨어에서 평균 8.5%의 복호화 속도향상을 보였으며, 보간 필터의 수행 시간을 평균 24.8% 향상시켰다.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2002.05c
• /
• pp.215-218
• /
• 2002

영상 내에 존재하는 객체를 배경에서 분리해내는 영상분할에 대한 연구는 일반적으로 픽셀중심, 에지기반, 영역기반 그리고 모델기반의 영역에서 이루어져왔다. Active Contour 모델은 객체를 영상에서 분리하는 에지기반의 영상분할 방식이다. 전통적인 의미의 Active Contour 모델에서 사용한 그라디언트 함수 기반의 영상추출은 잡영이 많고 객체와 배경간 뚜렷한 경계가 없는 객체를 검출하는데는 그 한계를 보이고 있다. 이런 한계를 극복하고자 제안된 방법이 Mumford-Shah equation과 Lipshitz 함수를 이용한 Chan과 Vese의 Active Contour Model이다. 그런데 이 모델은 잡영이 많고 경계선이 뚜렷하지 않은 영상을 분할하는데는 효과적이나, 불균형적 조명이 있는 영상에서 객체를 분리해 내는데는 한계를 보이고 있다. 본 논문은 이러한 단점을 극복하기 위해 불균형적인 영상을 균일화하는 방법을 Chan과 Vese의 Active Contour 방식을 적용하기 전에 적용 시켜 영상 내 객체를 보다 효과적으로 추출하는 방법을 제안한다.

• Journal of the Korean Society of Radiology
• /
• v.9 no.4
• /
• pp.227-234
• /
• 2015

In this paper, we studied possibility about application for CAD on diffuse liver disease through pixel texture analysis parameters(average gray level, skewness, entropy) which based statistical property brightness histogram and image analysis using brightness difference liver and kidney parenchyma. The experiment was set by ROI ($50{\times}50$ pixels) on liver ultrasound images.(non specific, fatty liver, liver cirrhosis) then, evaluated disease recognition rates using 4 types pixel texture analysis parameters and brightness gap liver and kidney parenchyma. As a results, disease recognition rates which contained average brightness, skewness, uniformity, entropy was scored 100%~96%, they were high. In brightness gap between liver and kidney parenchyma, non specific was $-1.129{\pm}12.410$ fatty liver was $33.182{\pm}11.826$, these were shown significantly difference, but liver cirrhosis was $-1.668{\pm}10.081$, that was somewhat small difference with non specific case. Consequently, pixel texture analysis parameter which scored high disease recognition rates and CAD which used brightness difference of parenchyma are very useful for detecting diffuse liver disease as well as these are possible to use clinical technique and minimize reading miss. Also, it helps to suggest correct diagnose and treatment.