• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 추계종합학술대회 논문집(4)
• /
• pp.167-170
• /
• 2000

In this paper, we present a fast algorithm for the motion estimation using the efficient selection of an initial search position. In the method, we select the initial search position using the motion vector from the subsmpled images, the predicted motion vector from the neighbor blocks, and the (0,0) motion vector. While searching the candidate blocks, we use the spiral search pattern with the successive elimination algorithm(SEA) and the partial distortion elimination(PDE). The experiment results show that the complexity of the proposed algorithm is about 2∼3 times faster than the three-step search(TSS) with the PSNR loss of just 0.05［dB］∼0.1［dB］ than the full search algorithm PSNR. The search complexity can be reduced with quite a few PSNR loss by controling the number of the depth in the spiral search pattern.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2005년도 학술대회 논문집 정보 및 제어부문
• /
• pp.529-531
• /
• 2005

A block-based fast motion estimation algorithm is proposed in this paper to perform motion estimation based on the efficiently reduced search ranges in MPEG-4(ERS). This algorithm divides the search areas into several small search areas and the candidate small search area that has the lowest average of sum norm difference between current macroblock and candidate macroblock is chosen to perform block motion estimation using the Nobel Successive Elimination Algorithm (NSEA). Experimental results of the proposed algorithm show that the averaging PSNR improvement is better maximum 0.125 dB than other tested algorithms and bit saving effect is maximum 20kbps for some tested sequences in low-bit rate circumstance.

• Journal of Electrical Engineering and Technology
• /
• 제3권1호
• /
• pp.101-107
• /
• 2008

We introduce an optimization of fuzzy set-fuzzy systems based on IG (Information Granules). The proposed fuzzy model implements system structure and parameter identification by means of IG and GAs. The concept of information granulation was coped with to enhance the abilities of structural optimization of the fuzzy model. Granulation of information realized with C-Means clustering helps determine the initial parameters of the fuzzy model such as the initial apexes of the membership functions in the premise part and the initial values of polynomial functions in the consequence part of the fuzzy rules. The initial parameters are adjusted effectively with the help of the GAs and the standard least square method. To optimally identify the structure and the parameters of the fuzzy model we exploit GAs with successive tuning method to simultaneously search the structure and the parameters within one individual. We also consider the variant generation-based evolution to adjust the rate of identification of the structure and the parameters in successive tuning method. The proposed model is evaluated with the performance of the conventional fuzzy model.

• 한국해양공학회지
• /
• 제18권5호
• /
• pp.29-35
• /
• 2004

A new approach, referred to as a successive zooming genetic algorithm (SZGA), is proposed for identifying a global solution, using continuous zooming factors for optimization problems. In order to improve the local fine-tuning of the GA, we introduced a new method whereby the search space is zoomed around the design variable with the best fitness per 100 generation, resulting in an improvement of the convergence. Furthermore, the reliability of the optimized solution is determined based on the theory of probability, and the parameter used for the successive zooming method is optimized. With parameter optimization, we can eliminate the time allocated for deciding parameters used in SZGA. To demonstrate the superiority of the proposed theory, we tested for the minimization of a multiple function, as well as simple functions. After testing, we applied the parameter optimization to a truss problem and wicket gate servomotor optimization. Then, the proposed algorithm identifies a more exact optimum value than the standard genetic algorithm.

• 대한전자공학회논문지SP
• /
• 제40권5호
• /
• pp.388-396
• /
• 2003

본 논문에서 정수단위 화소(integer pixel unit)로 움직임 예측(motion estimation)을 수행하는 빠른 움직임 예측(fast motion estimation) 알고리즘을 제안한다. 제안하는 방법은, sum norm을 사용하여 가장 좋은 움직임 벡터를 찾아내는 연속 제거 기법(SEA : Successive Elimination Algorithm)을 기반으로 16×16블록에서는 전체 영역에 대해 검색을 하고 16×8, 8×16, 8×8블록에서는 16×16블록의 움직임 벡터로부터 그 주위 8개의 위치에서 가장 좋은 벡터를 구하고, 8×4, 4×8, 4×4블록은 8×8블록의 움직임 벡터로부터 그 주위 8개의 위치에서 벡터를 검색하여 그 중에서 가장 좋은 움직임 벡터를 찾아내는 것이다. 이러한 움직임 검색(motion search) 방법을 가변 크기 블록(16×16, 16×8, 8×16, 8×8, 8×4, 4×8, 4×4)으로 움직임 예측을 하는 H.264 부호기(encoder)에 적용하였다. 제안하는 검색 알고리즘을 계산 복잡도 측면에서 보면, 조기 종료가 적용 안 된 나선형으로 전체 영역을 검색(Spiral full search without early termination)하는 방법보다 23.8배가 빨라졌고, 4×4 블록들의 계층적 SAD(Sum of Absolute Difference)를 이용하는 빠른 움직임 예측 방식보다 4.6배의 속도증가를 보인다. 반면에 신호 대 잡음 비(PSNR : Peak Signal to Noise Ratio)는 0.1dB에서 0.4dB정도 떨어짐을 보인다.

• 대한기계학회논문집A
• /
• 제27권5호
• /
• pp.681-687
• /
• 2003

Terrain surfaces have to be modeled in very detail and wheel-surface contacting geometry must be well defined in order to obtain proper ground-reaction and friction forces fur realistic simulation of off-road vehicles. Delaunay triangulation is one of the most widely used methods in modeling 3-dimensional terrain surfaces, and the T-search is a relevant algorithm for searching resulting triangular polygons. The T-search method searches polygons in a successive order and may not allow real-time computation of off-road vehicle dynamics if the terrain is modeled with many polygons, depending on the computer performance used in the simulation. The dynamic T-search, which is proposed in this paper, combines conventional T-search and the concept of the dynmaic-window search which uses reduced searching windows or sets of triangular surface polygons at each frame by taking advantage of the information regarding dynamic charactereistics of a simulated vehicle. Numerical tests show improvement of searching speeds by about 5% for randomly distributed triangles. For continuous search following a vehicle path, which occurs in actual vehicle simulation, the searching speed becomes 4 times faster.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 하계종합학술대회 논문집(4)
• /
• pp.143-146
• /
• 2000

Motion Estimation/compensation(ME/MC) is one of the efficient interframe ceding techniques for its ability to reduce the high redundancy between successive frames of an image sequence. Calculating the blocking matching takes most of the encoding time. In this paper a new fast block matching algorithm(BMA) is developed for motion estimation and for reduction of the computation time to search motion vectors. The feature of the new algorithm comes from the center-biased checking concept and the trend of pixel movements. At first, Motion Vector(MV) is searched in ${\pm}$1 of search area and then the motion estimation is exploited in the rest block. The ASP and MSE of the proposed search algorithm show good performance.

• 한국정밀공학회지
• /
• 제16권1호통권94호
• /
• pp.116-124
• /
• 1999

In most conventional design optimization of dynamic system, design sensitivities are utilized. However, design sensitivities based optimization method has numbers of drawback. First, computing design sensitivities for dynamic system is mathematically difficult, and almost impossible for many complex problems as well. Second, local optimum is obtained. On the other hand, Genetic Algorithm is the search technique based on the performance of system, not on the design sensitivities. It is the search algorithm based on the mechanics of natural selection and natural genetics. GA search, differing from conventional search techniques, starts with an initial set of random solutions called a population. Each individual in the population is called a chromosome, representing a solution to the problem at hand. The chromosomes evolve through successive iterations, called generations. As the generation is repeated, the fitness values of chromosomes were maximized, and design parameters converge to the optimal. In this study, Genetic Algorithm is applied to the actual dynamic optimization problems, to determine the optimal design parameters of the dynamic system.

• 한국통신학회논문지
• /
• 제29권9C호
• /
• pp.1323-1329
• /
• 2004

본 논문은 고속 전역탐색법 중의 하나인 MSEA(multi level successive elimination algorithm)를 개선한 방식으로서, MSEA의 단계에 따른 norm의 계산 결과를 이용하여 최종 단계의 SAD를 예측함으로써 더 이상의 단계를 수행할 필요가 없다고 판단되는 단계의 계산을 생략함으로써 계산 량을 감소시키는 방법을 제안한다. 각 단계별 SAD의 예측을 위해 norm에 대한 이론적 분석이 이루어지며 실험을 통해 분석내용을 검증하고, 이를 바탕으로 새로운 알고리즘을 제안하고 실험을 통해 제안된 알고리즘의 성능을 평가한다.

• 한국시뮬레이션학회:학술대회논문집
• /
• 한국시뮬레이션학회 2002년도 추계학술대회 논문집
• /
• pp.163-170
• /
• 2002

Terrain surfaces have to be modeled in very detail and wheel-surface contacting geometry must be well defined in order to obtain proper ground-reaction and friction forces for realistic simulation of off-road vehicles. Delaunay triangulation is one of the most widely used methods in modeling 3-dimensional terrain surfaces, and T-search is a relevant algorithm for searching resulting triangular polygons. The T-search method searches polygons in successive order and may not allow real-time computation of off-road vehicle dynamics if the terrain is modeled with many polygons, depending on the computer performance used in the simulation. In order to accelerate the searching speed of T-search, a terrain database of triangular polygons is modeled in multi-levels by adopting the LOD (Level of Detail) method used in realtime computer graphics. Simulation results show that the new LOD search is effective in shortening the required computing time. The LOD search can be even further accelerated by introducing an NN (Neural Network) algorithm, in the cases where a appropriate range of moving paths can be predicted by cultual information of the simulated terrain, such as lakes, houses, etc.. Numerical tests show that LOD-NN search almost double the speed of the original T-search.