• The KIPS Transactions:PartB
• /
• v.17B no.1
• /
• pp.63-68
• /
• 2010

In this paper, we propose a multiresolution stereo matching method based on genetic algorithm using edge information. The proposed approach considers the matching environment as an optimization problem and finds the solution by using a genetic algorithm. A cost function composes of certain constraints which are commonly used in stereo matching. We defines the structure of chromosomes using edge pixel information of reference image of stereo pair. To increase the efficiency of process, we apply image pyramid method to stereo matching and calculate the initial disparity map at the coarsest resolution. Then initial disparity map is propagated to the next finer resolution, interpolated and performed disparity refinement. We valid our approach not only reduce the search time for correspondence but alse ensure the validity of matching.

• Journal of the Korea Computer Graphics Society
• /
• v.27 no.3
• /
• pp.25-29
• /
• 2021

초박형 라이트필드 카메라 시스템은 이미지 센서 위에 렌즈 어레이를 부착하는 방식으로 만들어진다. 이러한 초박형 라이트필드 카메라는 하나의 이미지 센서를 여러 개의 sub-aperture가 나눠쓰는 방식으로 되어있어 개별 이미지의 분해능이 낮으며, sub-aperture 이미지들을 융합해 추가적인 분해능 향상이 수행되어야 한다. 본 연구에서는 초박형 라이트필드 카메라 시스템을 개발했으며, 개발된 카메라 시스템을 위한 실시간 분해능 향상 알고리즘을 개발, 실험을 통해 검증했다. 개발된 초박형 라이트필드 카메라는 두께 2mm, 24개(6×4)의 551×551 해상도의 sub-aperture로 구성되어 있으며, 임베디드 컴퓨팅 보드를 사용해 휴대가 가능하도록 제작되었다. 실시간 분해능 향상 알고리즘은 임베디드 컴퓨팅 보드의 GPU에서 병렬처리를 통해 라플라시안 피라미드 기반의 이미지 융합 알고리즘을 수행한다. 실험을 통해 검증한 결과로, 개발 시스템은 MTF50값이 평균 35% 정도 개선되었으며, 10.65fps의 처리속도로 실시간 처리가 가능함을 확인했다.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.44 no.10
• /
• pp.853-861
• /
• 2016

This paper deals with a new technique utilizing the angular speed of the reaction wheels to determine attitudes and angular rates for gyroless satellites. The suggested algorithm in this study is designed to determine the precise attitude and angular rates under actual space environments by the support of the angular speeds of reaction wheels based on the extended Kalman filter. Furthermore, the proposed approach is also designed to estimate not only the attitude and angular rates of spacecraft but the external disturbances. The numerical simulation was conducted for gyloless spacecraft installed with four reaction wheels of the pyramid-type configuration. The performance of the proposed algorithm is verified by using numerical simulations.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.12
• /
• pp.176-183
• /
• 2013

We propose a rolled fingerprint merging algorithm that effectively merges plain fingerprints in consecutive frame units that are fed through rolling and detects more fingerprint minutiae in order to increase the fingerprint recognition rate. The proposed rolled fingerprint merging algorithm uses a adaptive projection mask; it contains a detector that separates plain fingerprints from the background and a projection mask generator that sequentially projects the detect ed images. In addition, in the merging unit, the pyramid-shaped projection method is used to detect merged rolled fingerprints from the generated variable projective mask, starting from the main images. Simulations show that the extracted minutia e are 46.79% more than those from plain fingerprints, and the proposed algorithm exhibits excellent performance by detecting 52.0% more good fingerprint minutiae that are needed for matching.

• Proceedings of the KSRS Conference
• /
• 2008.03a
• /
• pp.80-85
• /
• 2008

최근 도심지역이 급변하고 고해상도 위성영상의 보급이 증가함에 따라 고해상도 위성영상을 이용한 수치표고모델과 정사영상 생성에 관한 연구가 활발해 지고 있다. 본 연구에서는 IKONOS, SPOT5, QUICKBIRD, KOMPSAT2 위성영상을 이용하여 DEM 과 정사영상을 생성하였으며 USGS DTED 와 기준점을 이용하여 결과의 정확도를 비교 분석하였다. 보다 정확한 DEM 생성을 위해 자동 피라미드 알고리즘을 적용하고 영상 정합시 에피폴라 기하학을 적용하였다. 정사 영상 생성시 DTED 높이값을 이용하여 보정을 수행하였으며 생성 속도를 높이기 위하여 리샘플링 그리드를 적용하였다. 본 연구에서 DEM 과 정사영상 생성시 QUICKBIRD 와 SPOT5 의 경우 영상의 용량이 매우 커 메모리 부족문제와 알고리즘 수행 속도 저하가 발생함을 확인하였다. 이를 개선하기 위하여 DEM 생성시 정합 후보점의 개수를 줄이는 알고리즘을 고안하여 기존에 메모리 문제로 생성하지 못했던 QUICKBIRD와 SPOT5 의 DEM 을 생성하였으며 정사 영상 생성시 리샘플링 그리드를 적용하여 고해상도 정상영상 생성 속도 개선에 상당한 효과를 가져왔다. 그러나 고해상도 위성 영상의 용량이 점점 커져감에 따라 이러한 메모리 문제와 처리 속도 저하에 관한 문제는 추후 계속적으로 연구되어야 할 부분이라고 할 수 있다. 본 연구에서 생성한 IKONOS, SPOT5, QUICKBIRD DEM 의 정확도를 USGS DTED 와 비교한 결과 13${\sim}$15 m 정도의 RMS 높이 오차가 산출되었으며 생성된 IKONOS, QUICKBIRD, KOMPSAT2 정사영상을 기준점과 비교한 결과 3 m 정도의 거리오차가 산출되었음을 확인하였다.

• Journal of IKEEE
• /
• v.23 no.1
• /
• pp.235-241
• /
• 2019

In this paper, we propose a new algorithm to detect rear-approaching vehicle using the frame similarity of ROI(Region of Interest) based on deep learning algorithm for use in agricultural machinery systems. Since the vehicle detection system for agricultural machinery needs to detect only a vehicle approaching from the rear. we use Faster R-CNN model that shows excellent accuracy rate in deep learning for vehicle detection. And we proposed an algorithm that uses the frame similarity for ROI using constrained conditions. Experimental results show that the proposed method has a detection rate of 99.9% and reduced the false positive values.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.16 no.11
• /
• pp.1154-1166
• /
• 1991

Optical flow is important not only for determining velocity and trajectory of the object but also for image segmentation and three dimensional information. In this paper an algorithm for segmentation of the optical flow field is presented. This method is based on a pyramid linking method combined with the proposed models. In this method each node contained a model of the flow in the region that it represented regions were combined by taking the model that best fit the union of the two regions. Each node linked to one of its fathers based on the error between the pixels represented by the node and the father's model of its flow. A major problem which has emerged in conventional researchs on optical flow field is sensitive to noise the proposed method is relatively insensitive to noise at the result of computer simulation the pyramid algorithm proposed in this paper seem to have useful properties.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38A no.1
• /
• pp.1-9
• /
• 2013

This paper propose a human action recognition method that uses bag-of-features (BoF) based on CS-LBP (center-symmetric local binary pattern) and a spatial pyramid in addition to the random forest classifier. To construct the BoF, an image divided into dense regular grids and extract from each patch. A code word which is a visual vocabulary, is formed by k-means clustering of a random subset of patches. For enhanced action discrimination, local BoF histogram from three subdivided levels of a spatial pyramid is estimated, and a weighted BoF histogram is generated by concatenating the local histograms. For action classification, a random forest, which is an ensemble of decision trees, is built to model the distribution of each action class. The random forest combined with the weighted BoF histogram is successfully applied to Standford Action 40 including various human action images, and its classification performance is better than that of other methods. Furthermore, the proposed method allows action recognition to be performed in near real-time.

• Journal of the Korea Society of Computer and Information
• /
• v.4 no.3
• /
• pp.73-77
• /
• 1999

In this paper. wavelet transform coding for image conference is studied. Original video frames are transformed into hierarchical pyramidal images with multiresolution using the band property of wavelet transform coefficients. Moving information between neighboring frames is obtained from the low-resolution band. Also, to control the video coding procedure. a new picture set filter is proposed. This filter controls the compression ratio of each frame depending on the correlation to the reference frame by selectively eliminating less important high-resolution areas. Consequently. video quality can be preserved and bit rate can be controlled adaptively In the simulation, to test the performance of the proposed coding method, comparisons with the full search block matching algorithm and the differential image coding algorithm are made. Consequently. the proposed method shows a reasonably good performance over existing ones.

• The KIPS Transactions:PartB
• /
• v.8B no.3
• /
• pp.311-318
• /
• 2001

본 논문에서는 중첩되는 영상사이의 평면 투영 변환을 구하는 방법을 제안하였다. 제안한 방법은 정규상관과 직사각형-사변형 평면 투영 변환에 기반 한다. 블록 정합을 이용하여 전역 이동을 계산하고, 중첩되는 영역의 정규 상관 계수 값을 최대로 하는 4개의 대응점을 찾기 위하여 가우시안 영상 피라미드에서 SA(simulated annealing) 알고리즘을 사용하였다. 이들 대응점에서 직사각형-사변형으로의 사상을 이용하여 평면 투영 변환을 계산하고, 마지막으로 중첩되는 영역의 RGB 컬러 값을 선형 가중치에 의해 혼합하였다. 실험으로 세 장의 영상을 한 장읠 큰 모자익 영상으로 합성하는 결과를 보였다.