• Journal of the Korea Convergence Society
• /
• v.8 no.6
• /
• pp.29-36
• /
• 2017

In this paper, we propose a method for generating improved saliency map by learning multiple features to improve the accuracy and reliability of saliency map which has similar result to human visual perception type. In order to overcome the inaccurate result of reverse selection or partial loss in color based salient area estimation in existing salience map generation, the proposed method generates multi feature data based on learning. The features to be considered in the image are analyzed through the process of distinguishing the color pattern and the region having the specificity in the original image, and the learning data is composed by the combination of the similar protrusion area definition and the specificity area using the LAB color space based color analysis. After combining the training data with the extrinsic information obtained from low level features such as frequency, color, and focus information, we reconstructed the final saliency map to minimize the inaccurate saliency area. For the experiment, we compared the ground truth image with the experimental results and obtained the precision-recall value.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.05a
• /
• pp.336-337
• /
• 2013

전 세계적으로 3D영화 및 디스플레이 기기에 대한 일반인들의 관심이 증가하고 있다. 하지만 편안한 시청영역(Comfort Zone)에서도 3D를 인식하지 못하는 사람들도 적지 않지 않다. 이런 사람들은 "자신의 눈"에 이상이 있는 것도 모른 채 3D 콘텐츠를 멀리하게 되지만, 안과적 조치를 통해 3D영상 시청 시 어지럼증 유발을 상당부분 제거할 수 있다. 기존의 입체시 검사 도구는 왜곡 및 실감정도를 파악할 수가 없다. 이를 세밀하게 측정할 수 있는 3D 콘텐츠 깊이 측정기를 제작하고, 이에 적합한 콘텐츠를 제작하였다. 3D 콘텐츠 깊이 측정기는 기능을 제어하는 콘트롤 부와 콘텐츠를 보여주는 컨벌젼스 룸으로 구성되어 있으며, 장비를 통해 3D 오브젝트의 깊이나 돌출거리와 같은 여러 파라메타들을 측정할 수 있다. 콘텐츠 제작에 대해선, 오브젝트의 위치는 여섯 부분으로 분류하고, 양안시차(disparity)는 $+2^{\circ}{\sim}-3^{\circ}$ 범위에 $1^{\circ}$ 간격으로 6레벨로 구성하였다. 하나의 오브젝트에 대해 서른여섯개의 오브젝트를 제작하였고, 이들의 조합을 통해 다양한 콘텐츠를 구성하도록 했다. 이를 바탕으로, 안간거리 6.5cm를 대상으로 돌출거리를 측정하였다.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.13 no.6
• /
• pp.1250-1256
• /
• 1989

본 연구에서는 수치계산의 불안정성 때문에 과거 고차 정확도 유한차분법 들로 수치계산하기 어려웠던 날카로운 돌출부 위를 흐르는 난류유동현상을 ASQUICK 차분법으로 수치해석하고 그 수렴해를 얻음으로써 수치계산의 안정성을 확인하며 수치계산 결과를 HYBRID차분법에 의한 수치해석치 및 재순환난류유동의 특성치라 할 수 있는 재순환영역길이 실험치와 상호비교함으로써 수치계산의 정확도 향상효과를 확인하고자 한다. 더불어 수치계산의 수렴과정에 큰 영향을 미치는 것으로 잘 알려 져 있는 속도 압력교정법으로써의 PISO법을 ASQUICK차분법과 함께 사용해 PISO-ASQU- ICK조합의 유용성을 확인하고자 한다.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.20 no.2
• /
• pp.741-751
• /
• 1996

The coupled conduction and convection heat transfer from the protruding heat source in a vertical channel is numerically investigated. Conjugate solution of the two-dimensional energy equation is obtained for the incompressible air flow over the rectangular block with local heat source. It was found that several recirculation zones and separation bubble near the block were related to Re and Gr. And the results show that fractions of the heat transfer through each of the block face, maximum temperature of the block and the relative effect of each parameter on the maximum temperature and heat transfer.

• Proceedings of the KIEE Conference
• /
• 1999.07a
• /
• pp.182-184
• /
• 1999

회수에너지를 저장한 콘덴서전압을 이용하여 평활전류 구동방식에서 SRM이 전동기 영역으로 동작할 경우에는 콘덴서에 저장된 에너지 모두를 전동기영역에서 소모할 수 있으나, 발전모드가 장시간 유지되면 콘덴서의 에너지는 더 이상 저장할 수 없는 상태가 되는 치명적인 약점을 갖고 있다. 따라서 본 연구에서는 전동기 영역 및 발전기영역에서 평활전류 모드로 동작할 수 있는 인버터회로를 제안하고, 발전에너지를 효과적으로 사용할 수 있는 스위칭 방식을 돌출하였다.

• Journal of KIISE:Software and Applications
• /
• v.36 no.11
• /
• pp.912-919
• /
• 2009

A novel algorithm using 2D tensor voting and edge-based approach is proposed for text detection in natural scene images. The tensor voting is used based on the fact that characters in a text line are usually close together on a smooth curve and therefore the tokens corresponding to centers of these characters have high curve saliency values. First, a suitable edge-based method is used to find all possible text regions. Since the false positive rate of text detection result generated from the edge-based method is high, 2D tensor voting is applied to remove false positives and find only text regions. The experimental results show that our method successfully detects text regions in many complex natural scene images.

• Journal of Life Science
• /
• v.19 no.8
• /
• pp.1062-1066
• /
• 2009

N-Acetylglucosamine kinase (GlcNAc kinase or NAGK; EC 2.7.1.59) catalyzes the phosphorylation of GlcNAc to GlcNAc-6-phosphate (GlcNAc-6-P). Despite detailed characterization of the enzyme itself, there have been few studies on the expression of NAGK in mammalian tissues. In the rat hippocampal neuron in culture, NAGK-immunoreactivity (IR) formed clusters in somatodendritic domains. In this study we characterized the NAGK clusters that protrude out the long axis of dendritic shafts. By double-labeling of the neurons with antibodies against NAGK and various synaptic proteins, we show that NAGK is positioned at the base of spines, while there were no NAGK protrusions into inhibitory postsynaptic sites. Immunoblot analysis showed that NAGK was included in synaptosomes but not in PSD fractions. Our results indicate that the NAGK clusters at the dendritic periphery protrude into spines.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.05a
• /
• pp.248-251
• /
• 2016

This thesis has discussed how to recognize and convert raised letters on tire to data and collect such data. Unlike the existing recognition system, the system presented by this thesis recognizes raised letters on tire through detecting letters after converting the Kinect camera image into image data in the preprocessing stage. After then, numbers and letters are analyzed through image improvement by use of binary images, noise filter, etc. In the recognition stage, letter distinction is used and raised letters on tire are recognized 100% through correction of errors by way of the correction algorithm for tire data recognition errors. In this paper it will be the development of a method of recognizing characters and the tire technology. Although there are many ways to the already recognized characters, Tire characters requires a technique different from the more general character recognition. For this reason and to develop additional technical methods and algorithms for character recognition.

• Journal of Korea Water Resources Association
• /
• v.39 no.4 s.165
• /
• pp.313-320
• /
• 2006

In this study, hydraulic characteristics of thalweg line occurred near the single groyne and recirculation zone around downstream of the structure were discussed from the experiments which conducted in fixed flat-bed channel flume. The thalweg line is the stream line where the maximum velocity occurs and flow separation appeared at the tip of groyne changes the thalweg alignment. In this study, the variations in flow fields which were caused by different length and permeability of groyne were measured by LSPIV(Large Scale Particle Image Velocimetry), and also the characteristics of thalweg line and separation area were analyzed. From these results, it is found that length to thalweg line from the channel wall $T_{CL}$ and height of separation area $S_h$ vary the channel width up to 75%, 50% in the change of length and permeability of groyne, however the Froude number does not affect on $T_{CL}$ and $S_h$. Velocity along the thalweg $U_{CL}$ Increased by twice the mean velocity, and the maximum velocity occurred in downstream where the distance $5{\sim}6$ times of groyne length away from the groyne, which does not get affected by length and permeability of groyne and Froude number.

• The Journal of the Korea Contents Association
• /
• v.3 no.4
• /
• pp.17-22
• /
• 2003

In this paper, we propose the new side-view face detection method in color images which contain faces over one. It uses color and the geometrical distance between nose and chin. We convert RGB to YCbCr color space. We extract candidate regions of face using skin color information from image. And then, the extracted regions are processed by morphological filter, and the processed regions are labeled. Also, we correct the gradient of inclined face image using projected character of nose. And we detect the inclined side-view faces that have right and left 45 tips by within via ordinate. And we get 92% detection rate in 100 test images.