• 제어로봇시스템학회:학술대회논문집
• /
• 1993.10b
• /
• pp.70-75
• /
• 1993

We represent a method of 3-D object recognition using multi images in this paper. The recognition process is executed as follows. Object models as prior knowledgement are generated and stored on a computer. To extract features of a recognized object, three CCD cameras are set at vertices of a regular triangle and take images of an object to be recognized. By comparing extracted features with generated models, the object is recognized. In general, it is difficult to recognize 3-D objects because there are the following problems such as how to make the correspondence to both stereo images, generate and store an object model according to a recognition process, and effectively collate information gotten from input images. We resolve these problems using the method that the collation on the basis of features independent on the viewpoint, the generation of object models as enumerating some candidate models in an early recognition level, the execution a tight cooperative process among results gained by analyzing each image. We have made experiments based on real images in which polyhedral objects are used as objects to be recognized. Some of results reveal the usefulness of the proposed method.

• Journal of KIBIM
• /
• v.5 no.2
• /
• pp.46-54
• /
• 2015

Vision-based tracking has been proposed as a means to efficiently track a large number of construction resources operating in a congested site. In order to obtain 3D coordinates of an object, it is necessary to employ stereo-vision theories. Detecting and tracking of multiple objects require an entity matching process that finds corresponding pairs of detected entities across the two camera views. This paper proposes an efficient way of entity matching for tracking of construction workers. The proposed method basically uses epipolar geometry which represents the relationship between the two fixed cameras. Each pixel coordinate in a camera view is projected onto the other camera view as an epipolar line. The proposed method finds the matching pair of a worker entity by comparing the proximity of the all detected entities in the other view to the epipolar line. Experimental results demonstrate its suitability for automated entity matching for 3D vision-based tracking of construction workers.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.7
• /
• pp.1439-1446
• /
• 2011

This paper describes the algorithm that lowers the dimension, maintains the object recognition and significantly reduces the eigenspace configuration time by combining the higher correlation feature information and Principle Component Analysis. Since the suggested method doesn't require a lot of computation than the method using existing geometric information or stereo image, the fact that it is very suitable for building the real-time system has been proved through the experiment. In addition, since the existing point to point method which is a simple distance calculation has many errors, in this paper to improve recognition rate the recognition error could be reduced by using several successive input images as a unit of recognition with K-Nearest Neighbor which is the improved Class to Class method.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.11a
• /
• pp.363-366
• /
• 2009

변이영상은 두 스테레오 영상의 시차에 의해 발생하는 각 화소의 변위를 수록한 영상이다. 이 영상은 깊이영상을 생성하여 시점 간 가상영상을 생성하는데 사용된다. 따라서 변이영상은 다시점 비디오 서비스와 직접적인 연관이 있다. 본 논문에서는 유일성(uniqueness)제약과 순차성(ordering) 제약을 사용하여 기준영상과 참조영상 사이의 관계를 이용하여 생성한 변이 공간 영상(DSI : disparity space image)으로부터 비용 행렬을 계산하여 최적의 변이 경로를 찾아가는 다이내믹 프로그래밍을 분석 하였다. 다이내믹 프로그래밍은 정밀한 변이 맵을 얻을 수 있고, 다른 방식들에 비해 연산 속도가 빠르다는 장점을 가지고 있지만, 영상의 화소값의 변화가 없는 영역에서 이전의 경로를 계속 유지하려는 성질에 때문에 발생 하는 오류확산과 가려진 (occluded) 영역에 의한 오차로 인해 정확한 경로를 찾을 수 없는 경우가 빈번히 발생 하여 에러율이 높아지는 단점을 가지고 있다. 이러한 이론을 토대로 기존의 기법들에 비하여 정확도가 우수한 기법들을 제안하였다. 개선된 역 추적 과정을 이용하여 기존의 다이내믹 프로그래밍 기반의 스테레오 정합 기법들보다 우수성이 뛰어난 결과들을 나타내었다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2011.07a
• /
• pp.365-368
• /
• 2011

본 논문에서는 시청자의 색상 특성을 고려하여 양안식 영상에서의 색상 불일치를 예측하는 방법을 제안했다. 먼저 색상 불일치 예측을 위한 주관적 평가로 색상 민감도 평가를 수행했다. 기본 세 채널 색 R, G, B에 대해 각각의 색상의 오차 범위를 결정한 후, 이를 기반으로 관심도와 복잡도를 평가하여 보다 세밀한 주관적 평가를 수행했다. 실험 결과를 통해 제안한 방법이 기존의 색상의 품질을 평가하기 위해 사용하는 유클리디안 색 거리 계산 방법과 US & Canadian Government Printing Offices의 품질 수준 규격(Qulaity Level Specifications)을 이용한 방법보다 효율적으로 색상 불일치 여부를 평가함을 확인했다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.12
• /
• pp.7302-7308
• /
• 2014

Detecting target objects robustly in natural environments is a difficult problem in the computer vision and image processing areas. This paper suggests a method of robustly detecting target objects in the environments where reflection exists. The suggested algorithm first captures scenes with a stereo camera and extracts the line and corner features representing the target objects. This method then eliminates the reflected features among the extracted ones using a homographic transform. Subsequently, the method robustly detects the target objects by clustering only real features. The experimental results showed that the suggested algorithm effectively detects the target objects in reflection environments rather than existing algorithms.

• Journal of Electrical Engineering and information Science
• /
• v.3 no.2
• /
• pp.211-221
• /
• 1998

It is becoming apparent that stereo matching algorithms need much information from high level cognitive processes. Otherwise, conventional algorithms based on bottom-up control alone are susceptible to local minima. We introduce a system that consists of two levels. A lower level, using a usual matching method, is based upon the local neighborhood and a second level, that can integrate the partial information, is aimed at contextual matching. Conceptually, the introduction of bottom-up and top-down feedback loop to the usual matching algorithm improves the overall performance. For this purpose, we model the image attributes using a Markov random field (MRF) and thereupon derive a maximum a posteriori (MAP) estimate. The energy equation, corresponding to the estimate, efficiently represents the natural constraints such as occlusion and the partial informations from the other levels. In addition to recognition, we derive a training method that can determine the system informations from the other levels. In addition to recognition, we derive a training method that can determine the system parameters automatically. As an experiment, we test the algorithms using random dot stereograms (RDS) as well as natural scenes. It is proven that the overall recognition error is drastically reduced by the introduction of contextual matching.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.5
• /
• pp.63-69
• /
• 2004

This paper proposes a novel method for achieving pose-invariant face recognition using cylindrical model. On the assumption that a face is shaped like that of a cylinder, we estimate the object's pose and then extract the frontal face image via a pose transform with previously estimated pose angle. By employing the proposed pose transform technique we can increase the face recognition performance using the frontal face images. Through representative experiments, we achieved an increased recognition rate from 61.43% to 94.76% by the pose transform. Additionally, the recognition rate with the proposed method achieves as good as that of the more complicated 3D face model.

• Smart Structures and Systems
• /
• v.25 no.5
• /
• pp.531-541
• /
• 2020

This paper demonstrates the possibility of evaluating the safety of a building right after an earthquake using consumer-grade surveillance cameras installed in the building. Two cameras are used in each story to extract the time history of interstory drift during the earthquake based on camera calibration, stereo triangulation, and image template matching techniques. The interstory drift of several markers on the rigid floor are used to estimate the motion of the geometric center using the least square approach, then the horizontal interstory drift of any location on the floor can be estimated. A shaking table collapse test of a steel building was conducted to verify the proposed approach. The results indicate that the accuracy of the interstory drift measured by the cameras is high enough to estimate the damage state of the building based on the fragility curve of the interstory drift ratio. On the other hand, the interstory drift measured by an accelerometer tends to underestimate the damage state when residual interstory drift occurs because the low frequency content of the displacement signal is eliminated when high-pass filtering is employed for baseline correction.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.41 no.4
• /
• pp.91-100
• /
• 2004

In this paper, we propose the method that analyzes the depth of object using 2 images in the disparity algorithm. It also presents the design and implementation of the proposed method for a real time processing. The proposed system uses the vertical strip structure for calculating similar pixel numbers for the processing and converts the depth of object into gray scale images in order to be displayed on various display devices. The hardware using the proposed method is operating with 30 frames/sec and verified by using the Altera APEX 20K1000EBC652-3. The proposed method is also Implemented into It by using the Hynix 0.35${\mu}{\textrm}{m}$ CB35 ASIC library and 256PQFP package.