• The Journal of the Korea Contents Association
• /
• v.13 no.5
• /
• pp.28-36
• /
• 2013

Motion-compensated frame interpolation (MCFI) techniques in video signal processing have many application areas. Frame rate up-conversion (FRUC) or distributed video coding (DVC) technique needs an effective MCFI algorithm. For these applications, it is necessary to develop an effective post-processing technique to improve visual qualities or to reduce virtual channel noises, resulting in the reduced channel bit rate. This paper proposes a reliability evaluation method based on spatio-temporal characteristics for motion-compensated interpolated blocks. The proposed algorithm investigates the temporal matching characteristics for current frame and then is designed in such a way that it can measure temporal characteristics as well as the spatial ones. Through computer simulations, it is shown that the proposed method outperforms the conventional temporal matching method.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.14 no.6
• /
• pp.87-97
• /
• 2018

Rainfall depth is an important meteorological information. Generally, high spatial resolution rainfall data such as road-level rainfall data are more beneficial. However, it is expensive to set up sufficient Automatic Weather Systems to get the road-level rainfall data. In this paper, we proposed to use deep learning to recognize rainfall depth from road surveillance videos. To achieve this goal, we collected two new video datasets, and proposed a new deep learning architecture named Temporal and Spatial Segment Networks (TSSN) for rainfall depth recognition. Under TSSN, the experimental results show that the combination of the video frame and the differential frame is a superior solution for the rainfall depth recognition. Also, the proposed TSSN architecture outperforms other architectures implemented in this paper.

• IEIE Transactions on Smart Processing and Computing
• /
• v.3 no.5
• /
• pp.271-274
• /
• 2014

Bayesian based Multi-Frame Super-Resolution (MF-SR) has been used as a popular and effective SR model. On the other hand, the texture region is not reconstructed sufficiently because it works on the spatial domain. In this study, the MF-SR method was extended to operate on the frequency domain to improve HF information as much as possible. For this, a spatially weighted bilateral total variation model was proposed as a regularization term for a Bayesian estimation. The experimental results showed that the proposed method can recover the texture region more realistically with reduced noise, compared to conventional methods.

• International Journal of Contents
• /
• v.4 no.2
• /
• pp.24-28
• /
• 2008

In this paper, we propose a new approach for content-based video retrieval using non-parametric based motion classification in the shot-based video indexing structure. Our system proposed in this paper has supported the real-time video retrieval using spatio-temporal feature comparison by measuring the similarity between visual features and between motion features, respectively, after extracting representative frame and non-parametric motion information from shot-based video clips segmented by scene change detection method. The extraction of non-parametric based motion features, after the normalized motion vectors are created from an MPEG-compressed stream, is effectively fulfilled by discretizing each normalized motion vector into various angle bins, and by considering the mean, variance, and direction of motion vectors in these bins. To obtain visual feature in representative frame, we use the edge-based spatial descriptor. Experimental results show that our approach is superior to conventional methods with regard to the performance for video indexing and retrieval.

• Proceedings of the KIEE Conference
• /
• 2005.05a
• /
• pp.167-169
• /
• 2005

The occurrence of a single bit error in transmission bitstream leads to serious temporal and spatial errors. Because moving picture coding as MPEG-2 based on block coding algorithm uses variable length coding and motion compensation coding algorithm. In this paper, we propose algorithm to conceal occurred error of I-frames in transmission channel using data of the neighboring blocks in decoder. We divide a damaged macroblock of I-frame into four sub blocks and compose new macroblock using the neighboring blocks for each sub block. We estimate the block with minimum difference value through block matching with previous frame for new macroblocks and replace each estimated block with damaged sub block in the same position. Through simulation results, the proposed algorithm will be applied to a characteristic of moving with effect and shows better performance than conventional error concealment algorithms from visual and PSNR of view.

• Proceedings of the Korean Association of Geographic Inforamtion Studies Conference
• /
• 2003.04a
• /
• pp.135-138
• /
• 2003

레이더 interferometry 기법을 이용하여 1995년 12월 11일과 12일에 획득된 ERS-1(orbit: 23180, frame 2853), ERS-2(orbit: 03507, frame: 2853) tandem 자료로부터 서해안(강화도-아산만) 지역의 조간대 DEM을 추출하였다. 사용된 SAR 자료는 영상획득 당시의 낮은 조위(ERS-1: 0.89 cm, ERS-2: -0.02 cm), 짧은 시간 간격(1일), 그리고 비교적 긴 기선 간격(270 m 수직 baseline)을 갖기 때문에 조간대 DEM 추출에 매우 적합하였다. 비록 추출된 DEM의 정확도 평가는 수행되지 않았지만 기존에 광학영상으로부터 추출된 남양만 지역의 waterline과 시각적으로 단순 비교될 수 있었다. 보다 정밀한 정확도 평가는 본 연구에서 사용한 자료와 같은 날 획득된 곰소만 지역의 자료(frame: 2886)를 사용하여 waterline 방법으로 추출된 DEM 및 현장 측량자료를 이용하여 수행될 것이다. Temporal 그리고 baseline decorrelation 등을 무시하고 SAR 영상의 위상측정 오차만을 이용하여 계산된 고도 정밀도는 SNR 20 dB에서 약 0.57 m이다.

• Computational Structural Engineering
• /
• v.7 no.1
• /
• pp.125-134
• /
• 1994

Response variability of reinforced concrete frame subjected to material property randomness has been evaluated with the aid of the finite element method. The spatial variation of Young's modulus is assumed to be a two-dimensional homogeneous stochastic process. Young's Modulus of concrete material has been investigated based on the uiaxial strength of concrete cylinder. Direct Monte Carlo simulation method is used to investigate the response of reinforced concrete frame due to the variation of Young's modulus with the Neumann expansion method and the pertubation method. The results by three analytic methods are compared with those by deterministic finite element analysis. These stochastic technique may be an efficient tool for evaluating the structural safety and reliability of reinforced concrete structures.

• Journal of Information Processing Systems
• /
• v.7 no.1
• /
• pp.103-110
• /
• 2011

In this paper, an adaptive motion vector smoothing scheme based on weighted vector median filtering is proposed in order to eliminate the motion outliers more effectively for improving the quality of side information in frame-based distributed video coding. We use a simple motion vector outlier reliability measure for each block in a motion compensated interpolated frame and apply weighted vector median filtering only to the blocks with unreliable motion vectors. Simulation results show that the proposed adaptive motion vector smoothing algorithm improves the quality of the side information significantly while maintaining low complexity at the encoder in frame-based distributed video coding.

• Journal of Biomedical Engineering Research
• /
• v.22 no.6
• /
• pp.527-534
• /
• 2001

Conventional stereo image acquisition uses a pair of frame grabbers in the CAS(Computer Assisted Surgery) system. In this Paper, we developed a synchronized stereo image acquisition method with only one frame grabber Two images from left and right camera each other. were merged with different color space without time delay and thus only one frame grabber was enough toy stereo image. Due to this synchronous Property of image acquisition, we can improve spatial revolution on the computation of 3D Position. Furthermore the overall costs for 3D navigator can be down and the extraction time of stereo Position tan be shortened.

• Journal of the korean Society of Automotive Engineers
• /
• v.12 no.2
• /
• pp.29-42
• /
• 1990

Three dimensional frame structures with such nonlinearities as large displacements, medium rotations, plastic hinges and local defects are efficiently analyzed by introducing the nonlinear rotational spring. Formulations are based on the incremental updated Lagrangian descriptions and the virtual work principle, Axial displacement and twisted angle in beam elements are interpolated linearly, while bending displacements are approximated by the Hermite polynomials. The modified are length method is used as a solution method. The moment-angle of rotation relationship obtained analytically or experimentally can be easily incorporated into the solution procedure. Several examples tested show that the present method can be used efficiently in analyzing nonlinear frame structures with plastic hinges or local defect.