• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.35 no.11
• /
• pp.1429-1436
• /
• 2011

In this paper, a new tube-radius-measuring algorithm has been proposed for effectively measuring the radii of small tubes under severe noise conditions that can also perform well when metal scraps that make it difficult to measure the radius correctly are inside the tube hole. In the algorithm, we adopt a fuzzy searching method that searches for the center of the inner circle by using fuzzy parameters for distance and orientation from the initial search point. The proposed algorithm has been implemented and tested on both synthetic and real-world tube images, and the performance is compared to existing circle-detection algorithms, such as the Hough transform and RANSAC methods, to prove the accuracy and effectiveness of the algorithm. From this comparison, it is concluded that the proposed algorithm has excellent performance in terms of measurement accuracy and computation time.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.49 no.2
• /
• pp.8-17
• /
• 2012

In this paper we proposes a new framework to obtain 3D shape information of thin film rapidly. The conventional equipments based on reflectometry are not suitable for obtaining 3D overall shape information of thin film rapidly since they require more than 30 minutes to measure the absolute thickness for 170 points. The proposed framework is based on an image analysis method that extracts contour lines from interference fringes images using Canny edge detector. The absolute thickness for contour lines are measured and then a height map from the contour lines is obtained by interpolation using Borgefors distance transformation. The extracted height map is visualized using the DirectX 3D terrain rendering method. The proposed framework can provide 3D overall shape information of thin film in about 5 minutes since relatively small number of real measurement for contour lines is required.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2014.11a
• /
• pp.882-885
• /
• 2014

이미지 혹은 영상에 대한 자동 태깅은 해당 콘텐츠에 대한 추가적인 정보를 자동으로 시스템에 제공하는 기술로써 영상 인식, 콘텐츠 매시업, 정보 검색 등 다양한 기술/서비스 분야에서 여러 목적으로 활용되고 있다. 특히, 방송 콘텐츠는 많은 양의 정보를 제한된 영역 및 시간에 축약하여 담고 있기 때문에 영상 처리 기술을 통한 객체 인식이나, 콘텐츠 매시업, 추천 서비스 등의 성능 향상을 위해 자동 혹은 수동 태깅을 통한 정보 제공이 요구된다. 본 논문에서는 블로그를 이용한 프레임 단위의 방송 콘텐츠 태깅 기술을 제안한다. 제안하는 기술은 기존의 콘텐츠 단위의 정보 제공이나, 수동 태깅 된 정보를 제공하는 기술들과 달리, 영상의 각 프레임에 대한 자동 태깅을 목표로 한다. 제안하는 방법은 거리 학습을 통해 영상의 각 프레임이 가지는 특성을 고려한 모델을 구축한 후, 이를 토대로 영상의 프레임들과 블로그의 이미지를 매칭한다. 매칭된 결과를 기반으로 특정 블로그는 영상 내 특정 프레임 구간에 태깅 된다. 제안한 방법은 이미지 매칭 성능을 측정하여 평가하였다. 블로그 이미지에 대해 Top 1 매칭 프레임을 살펴본 결과, 70%의 정확률을 보였다. 소프트 매칭(Top n)의 경우, 최대 90%의 성능을 얻을 수 있음을 실험을 통해 알 수 있었다.

• Journal of Broadcast Engineering
• /
• v.25 no.1
• /
• pp.83-93
• /
• 2020

With the rapid increase of vehicles, various traffic problems, e.g., car crashes, traffic congestions, etc, frequently occur in the road environment of the urban area. To overcome such traffic problems, intelligent transportation systems have been developed with a traffic flow analysis. The traffic flow, which can be estimated by the vehicle counting scheme, plays an important role to manage and control the urban traffic. In this paper, we propose a novel vehicle counting method based on predicted centers of each lane. Specifically, the centers of each lane are detected by using the accumulated movement of vehicles and its filtered responses. The number of vehicles, which pass through extracted centers, is counted by checking the closest trajectories of the corresponding vehicles. Various experimental results on road CCTV videos demonstrate that the proposed method is effective for vehicle counting.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.7
• /
• pp.291-296
• /
• 2015

In this paper, we propose a new confidence measure using pixel selection for eye detection and design a hybrid eye detector. For this, we produce sub-images by applying a pixel selection method to the eye patches and construct the BDA(Biased Discriminant Analysis) feature space for measuring the confidence of the eye detection results. For a hybrid eye detector, we select HFED(Haar-like Feature based Eye Detector) and MFED(MCT Feature based Eye Detector), which are complementary to each other, as basic detectors. For a given image, each basic detector conducts eye detection and the confidence of each result is estimated in the BDA feature space by calculating the distances between the produced eye patches and the mean of positive samples in the training set. Then, the result with higher confidence is adopted as the final eye detection result and is used to the face alignment process for face recognition. The experimental results for various face databases show that the proposed method performs more accurate eye detection and consequently results in better face recognition performance compared with other methods.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.203-204
• /
• 2022

본 논문에서는 다수의 적외선 카메라의 2D 패시브마커 영상을 이용한 3차원 리지드 바디(Rigid Body) 자세추정 방법을 제안한다. 1차로 개별 카메라의 내부 변수를 구하기 위해 체스보드를 이용한 칼리브레이션 과정을 수행하고, 2차 보정 과정에서 3개의 적외선 마커가 있는 삼각형 구조물을 모든 카메라가 관찰 가능하도록 움직인 후 프레임별 누적된 데이터를 계산하여 카메라 간의 상대적인 위치정보의 보정 및 업데이트를 진행한다. 이 후 각 카메라의 좌표계를 3D월드 좌표계로 변환하는 과정을 통해 3개 마커의 3차원 좌표를 복원하여 각 마커간 거리를 계산하여 실제 거리와의 차이를 비교한 결과 1mm 내외의 오차를 측정하였다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.9 no.9
• /
• pp.1035-1042
• /
• 2014

Object recognition is one of the key technologies of the monitoring system for the prevention of various intelligent crimes. The height is one of the physical information of a person, and it may be important information for identification of the person. In this paper, a method which can detect pedestrians from CCTV images and estimate the height of the detected objects, is proposed. In this method, GMM (Gaussian Mixture Model) method was used to separate the moving object from the background and the pedestrian was detected using the conditions such as the width-height ratio and the size of the candidate objects. The proposed method was applied to the CCTV video, and the height of the pedestrian at far-distance, middle- distance, near-distance was estimated for the same person, and the accuracy was evaluated. Experimental results showed that the proposed method can estimate the height of the pedestrian as the accuracy of 97% for the short-range, 98% for the medium-range, and more than 97% for the far-range. The image sizes for the same pedestrian are different as the position of him in the image, it is shown that the proposed algorithm can estimate the height of pedestrian for various position effectively.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2012.07a
• /
• pp.447-450
• /
• 2012

본 논문은 최근 수요와 관심이 급증하고 있는 초고선명 4K-UHD 영상의 방송TV 서비스 도입시에 영상의 시청환경 및 신호규격 선정의 가이드라인을 제공하는 것을 목표로 HEVC 코덱(HM5.0 참조 소프트웨어)으로 부/복호화 한 4K-UHD 영상에 대한 주관적 화질평가를 결과를 제시한다. ITU-R BT.500의 권고에 따라 DSIS 방법으로 실험 하였으며, 원본 영상 대비 목표 비트율(18, 23, 36 Mbps), 컬러 포맷(YUV444, YUV420), 시청거리(0.75H, 1.5H) 변화에 따른 주관적 화질 평기를 수행하였다. 실험결과 시청거리가 가깝고 비트율이 18Mbps로 낮게 책정한 조건에서도 평균 의사 점수(MOS)는 4.65로서 피 실험자들은 화질 차이를 거의 느끼지 못하는 것으로 나다났다. 또한 동일한 비트율 조건에서 컬러 포맷을 YUV444와 YUV420에 대한 비교 실험에서 각각 MOS 값은 4.71과 4.81로 측정 되었으며, 평가자들은 채도(Chrominance)의 해상도 보다 전체적인 PSNR이 높을수록 높은 MOS 값을 주었다. 이는 동일 비트율 대비 YUV444보다 YUV420의 컬러 포맷이 더 높은 주관적 화질을 가짐을 의미한다. 본 실험을 통해 HEVC 코덱을 사용하여 현재의 방송 대역폭 조건에 UHD 영상도 충분히 고품질로 전송 가능하며, 영상의 컬러 포맷은 YUV420가 YUV444 방식보다 동일 비트율 대비 더 높은 화질을 가질 것이라고 예상 할 수 있다.

• Journal of KIISE:Software and Applications
• /
• v.36 no.8
• /
• pp.623-632
• /
• 2009

In this paper, we propose a pulmonary nodule registration for the tracking of lung nodules in sequential CT scans. Our method consists of following five steps. First, a translational mismatch is corrected by aligning the center of optimal bounding volumes including each segmented lung. Second, coronal maximum intensity projection(MIP) images including a rib structure which has the highest intensity region in baseline and follow-up CT series are generated. Third, rigid transformations are optimized by normalized average density differences between coronal MIP images. Forth, corresponding nodule candidates are defined by Euclidean distance measure after rigid registration. Finally, template matching is performed between the nodule template in baseline CT image and the search volume in follow-up CT image for the nodule matching. To evaluate the result of our method, we performed the visual inspection, accuracy and processing time. The experimental results show that nodules in serial CT scans can be rapidly and correctly registered by coronal MIP-based rigid registration and local template matching.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.9
• /
• pp.67-76
• /
• 2010

This paper proposes a image-based water level measurement method, which adapt to the ruler's surface condition. When the surface of a ruler is deteriorated by mud, drifts, or strong light reflection, the proposed method judges the pollution of ruler by comparing distance between two levels: the first one is the end position of horizontal edge region which keeps the pattern of ruler's marking, and the second one is the position where the sharpest drop occurs in the histogram which is construct using image density based on the axis of image height. If the ruler is polluted, the water level is a position of local valley of the section having a maximum difference between the local peak and valley around the second level. If the ruler is not polluted, the water level is detected as the position having horizontal edges more than 30% of histogram's maximum value around the first level. The detected water level is converted to the actual water level by using the mapping table which is construct based on the making of ruler in the image. The proposed method is compared to the ultrasonic based method to evaluate its accuracy and efficiency on the real situation.