• The KIPS Transactions:PartB
• /
• v.15B no.1
• /
• pp.9-16
• /
• 2008

In this paper, we present an algorithm for synthesizing intermediate view image from a stereoscopic pair of images. An image of multiview is need for people in order to easily recognize 3D image. However, if many cameras are use for that, not only does system get more complicated but also transmission rating cause a big trouble. Hence, stereo images are photograph and issue on the sending side and algorithm to generate several intermediate view image is able to be use on the receiving side. The proposed method is based on disparity space image. First of all, disparity space image that is depicted by the gap of pixel followed by disparity of stereo image is generated. Disparity map is made by utilizing disparity space image for searching for optimal disparity path then eventual intermediate view image is generated after occlusion region which does not match is processed. Experimental results illustrate the performance of the proposed technique and we obtained a high quality image of more than 30 dB PSNR.

• The Journal of the Korea Contents Association
• /
• v.5 no.5
• /
• pp.1-9
• /
• 2005

In this paper, we suggest multi colony interaction ant reinforcement learning model. This method is a hybrid of multi colony interaction by elite strategy and reinforcement teaming applying Temporal Difference(TD) learning to Ant-Q loaming. Proposed model is consisted of some independent AS colonies, and interaction achieves search according to elite strategy(Intensification, Diversification strategy) between the colonies. Intensification strategy enables to select of good path to use heuristic information of other agent colony. This makes to select the high frequency of the visit of a edge by agents through positive interaction of between the colonies. Diversification strategy makes to escape selection of the high frequency of the visit of a edge by agents achieve negative interaction by search information of other agent colony. Through this strategies, we could know that proposed reinforcement loaming method converges faster to optimal solution than original ACS and Ant-Q.

• Proceedings of the KIEE Conference
• /
• 2000.07a
• /
• pp.537-540
• /
• 2000

The Distribution System consist of many tie-line switches and sectionalizing switches, operated a radial type. When an outage occurs in Distribution System, outage areas are isolated by system switches, has to restored as soon as possible. At this time, system operator have to get a information about network topology for service restoration of outage areas. Therefore, the searching result of restorative topology has to fast computation time and reliable result topology for to restore a electric service to outage areas, equal to optimal switching operation problem. So, the problem can be defined as combinatorial optimization problem. The service restoration problem is so important problem which have outage area minimization, outage loss minimization. Many researcher is applying to the service restoration problem with various techniques. In this paper, advanced Dijkstra algorithm is applied to searching a restoration topology, is so efficient to searching a shortest path in graph type network. Additionally, fuzzy rules and operator are applied to overcome a fuzziness of correlation with input data. The present technique has superior results which are fast computation time and searching results than previous researches, demonstrated by example distribution model system which has 3 feeders, 26 buses. For a application capability to real distribution system, additionally demonstrated by real distribution system of KEPCO(Korea Electric Power Corporation) which has 8 feeders and 140 buses.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.4 no.2
• /
• pp.397-406
• /
• 2000

Genetic algorithm is used as a means of search, optimization md machine learning, its structure is simple but it is applied to various areas. And it is about an active and effective controller which can flexibly prepare for changeable circumstances. For this study, research about an action base system evolving by itself is also being considered. There is to have a problem that depended entirely on heuristic knowledge of expert forming membership function and control rule for fuzzy controller design. In this paper, for forming the fuzzy control to perform self-organization, we tuned the membership function to the most optimal using a genetic algorithm(GA) and improved the control efficiency by the self-correction and generation of control rules.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.16 no.1
• /
• pp.5-26
• /
• 2013

Since the optimized use of sonar systems for target tracking is a practical problem for naval operations, the measure of mission achievability is needed for preparing efficient sonar-maneuver tactic. In order to quantify the mission achievability or Measure Of Effectiveness(MOE) for given sonar-maneuver tactics, we developed and tested a simulation algorithm. The proposed algorithm for tracking is based on Measure Of Performance(MOP) for localization and tracking system of sonar against target. Probability of Detection(PD) using steering beam patterns referenced to the aspect angle of sonar is presented to consider the tracking-performance of sonar. Also, the integrated software package, named as Optimal Acoustic Search Path Planning(OASPP) is used for generating sonar-maneuver patterns and vulnerability analysis for a given scenario. Through simulation of a simple case for which the intuitive solution is known, the proposed algorithm is verified.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2006.05a
• /
• pp.784-787
• /
• 2006

In this paper, we propose a method for nonlinear grapheme segmentation in Korean printed character type classification. The characters are subdivided into six types based on character type information. The feature vector is consist of mesh features, vertical projection features and horizontal projection features which are extracted from gray-level images. We classify characters into 6 types using Back propagation. Character segmentation regions are determined based on character type information. Then, an optimal nonlinear grapheme segmentation path is found using multi-stage graph search algorithm. As the result, a proposed methodology is proper to classify character type and to find nonlinear char segmentation paths.

• Korean Journal of Artificial Intelligence
• /
• v.12 no.2
• /
• pp.17-23
• /
• 2024

Various methods have been developed to predict the flight path of an air-launched weapon to intercept a fast-moving target in the air. However, it is also getting more challenging to predict the optimal firing zone and provide it to a pilot in real-time during engagements for advanced weapons having new complicated guidance and thrust control. In this study, a method is proposed to develop an optimized weapon engagement zone by the SCG (Scaled Conjugate Gradient) algorithm to achieve both accurate and fast estimates and provide an optimized launch display to a pilot during combat engagement. SCG algorithm is fully automated, includes no critical user-dependent parameters, and avoids an exhaustive search used repeatedly to determine the appropriate stage and size of machine learning. Compared with real data, this study showed that the development of a machine learning-based weapon aiming algorithm can provide proper output for optimum weapon launch zones that can be used for operational fighters. This study also established a process to develop one of the critical aircraft-weapon integration software, which can be commonly used for aircraft integration of air-launched weapons.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.4
• /
• pp.157-164
• /
• 2023

This paper suggests O(n²) polynomial time heuristic algorithm for corporate tax structure optimization problem that has been classified as NP-complete problem. The proposed algorithm constructs tax tree levels that the target holding company is located at root node of Level 1, and the tax code categories(T_e) 1,4,3,2 are located in each level 2,3,4,5 sequentially. To find the maximum tax-relief path from source(S) to target(T), firstly we connect the minimum witholding tax rate _minr_w(u, v) arc of node u point of view for transfer the profit from u to v node. As a result we construct the spanning tree from all of the source nodes to a target node, and find the initial feasible solution. Nextly, we find the alternate path with minimum foreign tax rate _minr_fi(u, v) of v point of view. Finally we choose the minimum tax-relief path from of this two paths. The proposed heuristic algorithm performs better optimal results than linear programming and Tabu search method that is a kind of metaheuristic method.

• Journal of Korean Society of Transportation
• /
• v.23 no.7 s.85
• /
• pp.99-110
• /
• 2005

Total travel cost of route which connects origin with destination (O-D) is consist of the total sum of link travel cost and route perception cost. If the link perception cost is different according to the origin and destination, optimal route search has limitation to reflect the actual condition by route enumeration problem. The purpose of this study is to propose optimal route searching formulation and algorithm which is enable to reflect different link perception cost by each route, not only avoid the enumeration problem between origin and destination. This method defines minimum unit of route as a link and finally compares routes using link unit costs. The proposed method considers the perception travel cost at both origin and destination in optimal route searching process, while conventional models refect the perception cost only at origin. However this two-way searching algorithm is still not able to guarantee optimum solution. To overcome this problem, this study proposed an orign based optimal route searching method which was developed based on destination based optimal perception route tree. This study investigates whether proposed numerical formulas and algorithms are able to reflect route perception behavior reflected the feature of origin and destination in a real traffic network by the example research including the diversity of route information for the surrounding area and the perception cost for the road hierarchy.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.3 no.1 s.4
• /
• pp.31-44
• /
• 2004

The study of Estimation model for the short-term travel time prediction. There is a different solution which has predicted the link travel time to solve this problem. By using this solution, the link travel time is predicted based on link conditions from time to time. The predicated link travel time is used to search the shortest path. Before providing a dynamic shortest path finding, the prediction model should be verified. To verify the prediction model, three models such as Kalman filtering, Stochastic Process, ARIMA. The ARIMA model should adjust optimal parameters according to the traffic conditions. It requires a frequent adjustment process of finding optimal parameters. As a result of these characteristics, It is difficult to use the ARIMA model as a prediction. Kalman Filtering model has a distinguished prediction capability. It is due to the modification of travel time predictive errors in the gaining matrix. As a result of these characteristics, the Kalman Filtering model is likely to have a non-accumulative errors in prediction. Stochastic Process model uses the historical patterns of travel time conditions on links. It if favorably comparable with the other models in the sense of the recurrent travel time condition prediction. As a result, for the travel time estimation, Kalman filtering model is the better estimation model for the short-term estimation, stochastic process is the better for the long-term estimation.