• The Journal of Korean Institute of Communications and Information Sciences
• /
• 제29권8C호
• /
• pp.1175-1183
• /
• 2004

This paper propose a method to trace a moving object based on the information directly obtained from MPEG-2 compressed video stream without decoding process. In the proposed method, the motion flow is constructed from the motion vectors involved in compressed video and then we calculate the amount of pan, tilt, zoom associated with camera operations using generalized Hough transform. The local object motion can be extracted from the motion flow after the compensation with the parameters related to the global camera motion. The moving object is designated initially by a user via bounding box. After then automatic tracking is performed based on the mean-shift algorithm of the motion flows of the object. The proposed method can improve the computation speed because the information is directly obtained from the MPEG-2 compressed video, but the object boundary is limited by blocks rather than pixels.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.177.3-177
• /
• 2001

In this paper, the design of the in-motion alignment system of Strapdown Inertial Navigation System(SDINS) using Robust Extended Kalman Filter(REKF) is presented. The compensation of errors in the aided navigation system is accomplished by the indirect feedback filtering. The performance of the aided navigation algorithm is very sensitive to the accuracy of the initial estimate, which is the characteristic of the EKF. Unfortunately, the initial attitude error can be very large during the in-motion alignment. To overcome the in-motion alignment under large initial attitude error problem, the REKF using linear robust filtering technique is proposed. The linear robust H$_2$ filter can be adopted for nonlinear ...

• The Journal of Korean Institute of Communications and Information Sciences
• /
• 제28권2C호
• /
• pp.193-200
• /
• 2003

In this paper, we propose a motion estimation method using hierarchical triangulation that changes the triangular mesh structure according to it's motion activity. The subdivision of triangular mesh is performed from the amount of motion that is calculated from the variance of image difference. As a result, node distribution is concentrated on the region of high activity. The subdivision method that makes it possible to yield hierarchical triangular mesh is proposed as well as the additional information reduction coding method for hierarchical mesh structure is described. By the simulation, proposed method have better performance than the conventional BMA(Block Match Algorithm) and the other mesh based method.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• 제46권5호
• /
• pp.120-125
• /
• 2009

For effective noise removal prior to video processing, noise power or noise variance of an input video sequence needs to be found exactly, but it is actually a very difficult process. This paper presents an accurate noise variance estimation algorithm based on motion compensation between two adjacent noisy pictures. Firstly, motion estimation is performed for each block in a picture, and the residue variance of the best motion-compensated block is calculated. Then, a noise variance estimate of the picture is obtained by adaptively averaging and properly scaling the variances close to the best variance. The simulation results show that the proposed noise estimation algorithm is very accurate and stable irrespective of noise level.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• 제32권4C호
• /
• pp.396-400
• /
• 2007

This paper proposes the fast partial difference elimination (PDE) algorithm. When the conventional PDE cannot skip the rest of matching procedure in a candidate block using a partial matching error, the proposed algorithm estimates to skip it again using the block matching error predicted from the computed partial matching error. The proposed algorithm can eliminate impossible candidate blocks earlier than the conventional PDE since the predicted block matching error is always bigger than the partial matching error. The simulation results show that the proposed algorithm can significantly reduce the computations while keeping image quality as good as the conventional PDE.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• 제12권7호
• /
• pp.1187-1192
• /
• 2008

This paper proposes a fast motion estimation using an adaptive search region and a new three step search. The proposed method improved in the quality of motion compensation image as $0.43dB{\sim}2.19dB$, according as it predict motion of current block from motion vector of neigher blocks, and adaptively set up search region using predicted motion information. We show that the proposed method applied a new three step search pattern is able to fast motion estimation, according as it reduce computational complexity per blocks as $1.3%{\sim}1.9%$ than conventional method.

• Journal of Korea Multimedia Society
• /
• 제8권3호
• /
• pp.345-354
• /
• 2005

Since motion estimation and motion compensation methods remove the redundant data to employ the temporal redundancy in images, it plays an important role in digital video compression. Because of its high computational complexity, however, it is difficult to apply to high-resolution applications in real time environments. If we have a priori knowledge about the motion of an image block before the motion estimation, the location of a better starting point for the search of an exact motion vector can be determined to expedite the searching process. In this paper presents the motion detection algorithm that can run robustly about recusive motion. The motion detection compares and analyzes two frames each other, motion of whether happened judge. Through experiments, we show significant improvements in the reduction of the computational time in terms of the number of search steps without much quality degradation in the predicted image.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• 제46권5호
• /
• pp.135-142
• /
• 2009

Since hold-type display systems have been developed, frame-rate up conversion (FRUC) is an essential technique to improve the temporal resolution in the display. FRUC improves the temporal resolution by interpolating one or multiple intermediate frames between two adjacent frames. In this paper, a new frame-rate up-conversion algorithm based on bilateral motion refinement with rotation is proposed. First, we perform bi-directional motion estimation between adjacent two frames to obtain a motion vector for each block. Then, we apply a modified median filtering to motion vectors for outlier-rejection and motion field smoothing. The filtered motion vectors are updated by the bilateral motion refinement with rotation. After the refined motion vector is obtained, the intermediate frame is generated by applying the overlapped block motion compensation (OBMC). Experimental results show that the proposed algorithm provides a better performance than the previous methods subjectively and objectively.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• 제33권11C호
• /
• pp.917-924
• /
• 2008

Motion estimation and compensation in the spatial domain check the searching area of specified size in the previous frame and search block to minimize the difference with current block. When we check the searching area, it consumes the most encoding times due to increasing the complexity. We can solve this fault by means of motion estimation using shifting matrix in the transform domain instead of the spatial domain. We derive so the existed shifting matrix to a new recursion equation that we decrease more computations. We modify simply vertical shifting matrix and horizontal shifting matrix in the transform domain for motion estimation of half-pixel accuracy. So, we solve increasing computation due to bilinear interpolation in the spatial domain. Simulation results prove that motion estimation by the proposed algorithm in DCT-based transform domain provides higher PSNR using fewer bits than results in the spatial domain.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• 제37권3호
• /
• pp.1-10
• /
• 2000

This paper proposes a novel video coding system with rate control based on fractal algorithm To overcome the demerits of excessive amounts of coded bit generated by previous fractal coding methodology. the proposed system classifies the Image into three classes such as background, motion compensation, and fractal coding area. The motion vector for motion compensation, and the fractal offset value that is difference value between the predicted offset and the least-square approximated value are coded with variable length code The decision method which determines threshold value of partitioning quadtree is applied to the bit-rate control algorithm considering the quantity of currently generated bits and fixed channel bandwidth Experimental result shows that the proposed system enhances compression ratio 1.8 times higher than previous method for the same image quality, and performs efficient rate control for fixed channel bandwidth.