• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2012.11a
• /
• pp.33-36
• /
• 2012

변이 탐색 범위는 스테레오 정합을 위한 변이 추정 알고리즘의 성능에 가장 큰 영향을 줄 수 있는 변수 중 하나이다. 일반적으로 이러한 변이 탐색 범위는 영상 전체의 최소 변이 값과 최대 변이 값의 범위로 정의된 전역적인 탐색 범위를 사용하게 된다. 이러한 전역적인 탐색 범위는 픽셀에 따라서 불필요한 범위에까지 탐색하게 하며, 이는 변이 추정의 성능적, 속도적 측면에서 좋지 않은 영향을 미치게 된다. 본 논문에서는 픽셀에 따라 변이 탐색 범위를 제한시키는 방법을 제안한다. 낮은 계산 복잡도를 갖는 간단한 변이 추정 알고리즘을 통해 먼저 변이 지도를 생성하여 각 픽셀 당 윈도우를 대응하여 변이 히스토그램을 만든다. 변이 히스토그램에 적응적인 임계 값을 적용하여 픽셀당 변이 후보군을 결정한다. 전역적인 변이 탐색 범위와 제안하는 방법으로 제한된 변이 탐색 범위를 각각 동일한 변이 추정 알고리즘에 적용해 보는 실험을 통해 제안하는 방법의 성능을 분석하였다. 실험 결과로부터 제안하는 방식은 변이 추정 알고리즘의 정확도 측면의 성능을 향상시켰고, 알고리즘의 속도를 평균 5배 이상 향상시켜주는 결과를 보여주었다.

• Journal of Internet Computing and Services
• /
• v.19 no.3
• /
• pp.43-48
• /
• 2018

In this paper, we propose an indoor-space recognition using DNN and super-pixel. In order to recognize the indoor space from the image, segmentation process is required for dividing an image Super-pixel is performed algorithm which can be divided into appropriate sizes. In order to recognize each segment, features are extracted using a proposed method. Extracted features are learned using DNN, and each segment is recognized using the DNN model. Experimental results show the performance comparison between the proposed method and existing algorithms.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.1
• /
• pp.457-461
• /
• 2005

본 논문에서는 RGB 컬러 정보와 오류 역전파 알고리즘을 이용한 신 차량 번호판 인식 방법을 제안한다. 차량 영상에서 평균 Blue값을 이용하여 차량 영상을 보정한다. 보정된 차량 영상에서 순수 Red픽셀과 현재 픽셀의 차이와 순수 Green 픽셀과 현재의 픽셀의 차이를 각각 구하여 Red 후보 영역과 Green 후보 영역으로 구분한다. 구분된 2개의 후보 영역의 픽셀 값을 오류 역전파 알고리즘에 적용하여 최종 Green 영역을 찾는다. 그리고 오류 역전파 알고리즘에 의해서 Green 영역으로 판명된 영역을 제외한 영역들은 잡음으로 처리한다. 잡음이 제거된 영역에 대해 수평 및 수직 히스토그램의 빈도수를 이용하여 번호판 영역을 추출한다. 추출된 번호판 영역에서 윤곽선 추적 알고리즘을 적용하여 개별 코드들을 추출하고, 오류 역전파 알고리즘을 적용하여 개별 코드들을 인식한다. 제안된 차량 번호판 추출 및 인식 방법의 성능을 평가하기 위하여 실제 비영업용 신 차량 번호판에 적용한 결과, 제안된 번호판 추출 방법이 기존의 HSI 정보를 이용한 번호판 추출 방법보다 추출률이 개선되었고 제안된 차량 번호판 인식 방법이 효율적인 것을 확인하였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.3
• /
• pp.531-536
• /
• 2009

In this paper, we propose $256{\times}256$ pixel array fingerprint sensor with an advanced circuits for detecting. The pixel level simple detection circuit converts from a small and variable sensing current to binary voltage out effectively. We minimizes an electrostatic discharge(ESD) influence by applying an effective isolation structure around the unit pixel. The sensor circuit blocks were designed and simulated in standard CMOS $0.35{\mu}m$ process. Full custom layout is performed in the unit sensor pixel and auto placement and routing is performed in the full chip.

• Journal of the Institute of Convergence Signal Processing
• /
• v.5 no.2
• /
• pp.119-125
• /
• 2004

The direction component uses mainly the sobel and FFT method. The sobel method is difficult to set representative direction when we wish to extract representative direction when we wish to extract representative direction component and the complicated processing process of sobel mask because same value appears in the low provision image or high provision image and we cannot accumulate tilt size in case of making accumulate after making unit vector to pixel. The method that uses FFT conversion for direction extraction is possible in case that ridge has correct direction specification and must use a special direction filter. After thinning the binary image to supplement above weak point in the paper, we extract direction component by pixel unit, and we extract the most direction components of pixel that exist in block of 8${\times}$8 pixels size as representative direction of ridge.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.47 no.10
• /
• pp.34-41
• /
• 2010

We proposed the new pixel compensation circuit with high aperture ratio and the driving method for the large-area, low-power AMOLED applications in this study. We designed with the low-temperature poly-silicon(LTPS) thin film transistors(TFTs) that has poor uniformity but good mobility and stability. To lower the error rate of the pixel circuit and to improve the aperture ratio for bottom emission method, we simplified the pixel compensation circuit. Because the proposed pixel compensation circuit with high aperture ratio has very low contrast ratio for conventional driving methods, we proposed the new driving method and circuit for high contrast ratio. Black data insertion was introduced to improve the characteristics for moving images. The pixel circuit was designed for 19.6" WXGA bottom-emission AMOLED panel, and the average aperture ratio of the pixel circuit is improved from 33.0% to 41.9%. For the TFT's $V_{TH}$ variation of ${\pm}0.2\;V$, the non-uniformity and contrast ratio of the designed panel was estimated under 6% and over 100000:1 respectively.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.4
• /
• pp.23-28
• /
• 2011

Ultrasound images for the abdominal muscles are complicated enough to have difficulty in interpreting their results. For better interpretation, magnifying the original image is necessary but its magnified image could be deteriorated and suffer from information loss. Thus, in this paper, we propose a magnifying method that reduces the gap between the original image and the magnified one in quality using a fuzzy method with weights for its brightness and interpolation. The proposed method extracts information of pixels in magnified image that have most similar characteristics of the original one by applying fuzzy membership function. In the process, the difference in the brightness between pixels of the magnified image and the original one using bilinear interpolation method and the weight value using the interpolation from multiplied values of four pixels are supplied to the fuzzy membership function. In this experiment, the proposed method reduces the cloudy phenomenon appears commonly compared to the bilinear interpolation method among those qualitative issues of image interpretation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.10
• /
• pp.2409-2416
• /
• 2015

This paper presents an orientation-based adaptive prediction method for effective lossless image compression. For a robust prediction, the proposed method estimates the directional information and the property near the current pixel in a support region-based fashion, not a pixel-based one which is sensitive to a small variation. We improve the prediction performance effectively by selection of the prediction pixel adaptively according to the similarity between support regions of the current pixel and the neighboring pixels. Comprehensive experiments demonstrate that the proposed scheme achieves excellent prediction performance measured in entropy of the prediction error compared to a number of conventional prediction methods such as MED, GAP, and EDP. Moreover the complexity of the proposed algorithm measured by average execution time is low compared to MED which is the simplest prediction method.

• Journal of the Korea Society of Computer and Information
• /
• v.12 no.4
• /
• pp.87-94
• /
• 2007

Imbalances in focus, luminance and color between stereo Pairs could cause disparity vector estimation error and increment of transmission data. If the distribution of errors in residual image is large, it may influence to lowering of compression performance. Therefore, in this paper, we propose an efficient balancing method between stereo pairs to reduce the effect. For this, we registrated stereo images using a FFT based method to consider the pixels in the occluded region, we eliminated the pixels of blocks which has large error of disparity vector estimation in balancing function estimation. The balancing function has estimated using histogram specification, local information of target image and residual image between stereo images. Experiments show that the proposed method is effective in error distribution, PSNR and disparity vector estimation. We expect that our method can be improving compression efficiency in stereo coding system.

• Journal of Sensor Science and Technology
• /
• v.12 no.4
• /
• pp.149-155
• /
• 2003

A PMOSFET photodetector for highly-sensitive active pixel sensor(APS) is presented. This sensor uses 5V power supply and has been designed and fabricated using I-poly and 2-metal $1.5{\mu}m$ CMOS technology. The feature of a PMOSFET photodetector is that the polysilicon gate of the PMOSFET was connected to n-well, in order to increase the photo sensitivity. The designed MOS photodetector has similar $I_{DS}-V_{DS}$ characteristics with a standard MOSFET. One dimensional image sensor with 16 pixels based on the PMOSFET photodetector has also been designed and fabricated. Unit pixel of the designed sensor consists of a PMOSFET photodetector and 4 NMOSFETs. Unit pixel area is $86{\mu}m{\times}90.5{\mu}m$ and its fill factor is about 12%.