• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.8 no.1
• /
• pp.127-135
• /
• 2009

This paper presents a new technique for corner shape based object retrieval from a database. The proposed feature matrix consists of values obtained through a neighborhood operation of detected corners. This results in a significant small size feature matrix compared to the algorithms using color features and thus is computationally very efficient. The corners have been extracted by finding the intersections of the detected lines found using Hough transform. As the affine transformations preserve the co-linearity of points on a line and their intersection properties, the resulting corner features for image retrieval are robust to affine transformations. Furthermore, the corner features are invariant to noise. It is considered that the proposed algorithm will produce good results in combination with other algorithms in a way of incremental verification for similarity.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.11b
• /
• pp.789-792
• /
• 2003

타부 탐색은 타부 전략 기법과 최급 강하 알고리즘이 결합된 알고리즘이다. 이는 한번 방문한 해는 다시 방문하지 않음으로써 지역 최적해에 수렴하지 않고 새로운 방향으로 움직이게 하여 공간 탐색 능력 효율을 높인다. 그러나 기존의 타부 탐색에서 이웃 해를 생성하는 방법에 따라 성능이 많이 좌우된다. 좋지 않은 이웃 해를 생성하는 탐색에서는 얻고자 하는 최적해에 수렴하는 시간이 많이 걸린다. 따라서 이웃 해를 생성할 때 해밍 거리를 고려하여 균형 있는 이웃 해론 생성하고, 해 공간은 탐색함으로써 우수한 최적해를 얻게 됨을 본 논문에서는 보여주고 있다. 이는 다양성도 보장되므로 최적해에 수렴해 가는 속도 또한 빠른 것을 보여주고 있다.

• Transactions of the Society of Information Storage Systems
• /
• v.1 no.2
• /
• pp.137-142
• /
• 2005

In this paper, we propose a new servo system design strategy to reduce the position error signal(PES) and track mis-registration(TMR) in magnetic disk drive systems. The proposed method provides a systematic design procedure based on the plant model and an optimal solution via an optimization with a 'Robust Random Neighborhood Search(RRNS)' algorithm. In addition, it guarantees the minimum PES level as well as stability to parametric uncertainties. Furthermore, the proposed method can be used to estimate the performance at the design stage and thus can reduce the cost and time for the design of the next generation product. The reduction of PES as well as robust stability is demonstrated by simulation and experiments.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2004.11a
• /
• pp.323-330
• /
• 2004

As mobile market grows more and more fast, the mobile contents market, especially music contents for mobile phones have recorded remarkable growth. In spite of this rapid growth, mobile web users experience high levels of frustration to search the desired music. New musics are very profitable to the content providers, but the existing collaborative filtering (CF) system can't recommend them. To solve these problems, we propose an extended CF system to reflect the user's real preference by representing the characteristics of users and musics in the feature space. We represent the musics using the music contents based acoustic features in multi-dimensional feature space, and then select a neighborhood with the distance based function. Furthermore, this paper suggests a recommendation for procedure for new music by matching new music with other users' preference. The suggested procedure is explained step by step with an illustration example.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.469-470
• /
• 2015

우리는 그룹핑 k-최근접 (Grouping k Nearest Neighbor; GkNN)질의를 지원하기 위하여 유연한 최근접객체(Flexible Nearest Neighbor; FNN)검색 방법을 제안한다. GkNN이란 기존에 제안된 kNN과 다르게 질의자가 요청한 k개의 객체를 모두 확인한 후에 이동 경로의 총합이 가장 작은 k개의 객체를 검색하는 방법이다. 기존 연구에서 제안된 최근접 객체들 (Nearest Neighborhood; NNH) 또한 이 문제를 해결하기 위하여 제안되었다. 그러나 NNH의 문제점은 객체 k와 p가 고정되어 있기 때문에 이동 환경에서 q에서 C까지의 거리가 증가하는 것이다. FNN의 환경은 NNH의 환경과 유사하다. 우리는 NNH의 q에서 집합 C 중 거리 중 가장 짧은 $c_i$ 선택한 후 q에서 $c_i$에 포함된 객체들 모두 검색하는 이동 경로의 총합과 FNN의 이동경로의 총 합을 비교하여 NNH의 문제점을 해결하였다.

• Journal of KIISE:Software and Applications
• /
• v.31 no.5
• /
• pp.570-576
• /
• 2004

It is very difficult to efficiently solve a large-scaled maximal covering problem(MCP) by a genetic algorithm. In this paper, we present new crossover and mutation operators specially designed for genetic algorithms to solve large-scaled MCPs efficiently. We also introduce a novel genetic algorithm employing unexpressed genes. Unexpressed genes are the genes which are not expressed and thus do not affect the evaluation of the individuals. These genes play the role of reserving information susceptible to be lost by the application of genetic operations but is suspected to be potentially useful in later generations. The genetic algorithm employing unexpressed genes enjoys the advantage of being able to maintain diversity of the population and thus can search more efficiently to solve large-scaled MCPs. Experiments with large-scaled real MCP data has shown that our genetic algorithm employing unexpressed genes significantly outperforms tabu search which is one of the popularly used local neighborhood search algorithms for optimization.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.9
• /
• pp.1193-1201
• /
• 2020

Cell planning, which determines the cell structure for location management of mobile terminals in mobile communications, has been dealt with as an important research task to determine network performance. Among the factors influencing the cell structure planning in mobile communication, the signal cost for location management plays the most important role. In this paper, we propose an optimization algorithm that minimizes the location management cost of all the cells used to plan the cell structure in the network with reporting cell structure in mobile communication. The proposed algorithm uses a Tabu search algorithm, which is a meta-heuristic algorithm, and the proposed algorithm proposes a new neighborhood generation method to obtain a result close to the optimal solution. In order to evaluate the performance of the proposed algorithm, the simulation was performed in terms of location management cost and algorithm execution time. The evaluation results show that the proposed algorithm outperforms the existing genetic algorithm and simulated annealing.

• The Journal of the Convergence on Culture Technology
• /
• v.6 no.1
• /
• pp.163-169
• /
• 2020

The purpose of this study is to find out whether the environment is based on the Color Universal Design (CUD, CUD Recommended Color Set Guidebook), which is located in the area with the highest proportion of elderly people in the city of Busan in Korea. Five nearby parks in the walking district of Busan were selected to measure the environmental color with a spectrophotometer(Minolta CM-2600d) and web color search. The range of the color survey was limited to the pavement; public facilities(sign, benches, restroom), and exercise&sports facilities. The results of the study are as follows. Overall, the Y color group was dominant, with similar color usage and intensity, and similar levels of saturation. This was analyzed based on the Color Universal Design theory, suggesting that the visual characteristics of the elderly were not taken into account, and the Neighborhood Park located in Busan was found to lack the environmental color of consideration for the visual characteristics. In addition, this study could act as a basic study that can be applied to social and environmental colors that reflect the color perception and cognitive characteristics of the elderly.

• Korean Security Journal
• /
• no.14
• /
• pp.109-123
• /
• 2007

Traffic terror is one of the terror types subject on transportation means. However, unlike the aircraft or ship, there is no case of negotiating with passengers as hostage but to unconditionally attack that the human casualty would be much greater. In act, the terror on subway or bus would have much bigger damage than the terror on aircraft or ship. Because of such fact, there is a need for full preparation thereto, however, there is no preparation of advance preventive measure, particularly, when comparing to the cases involving aircraft. In addition, the passengers who use railway and others in the position of citizens do not have any defense mechanism against any terror activity to make the problem even more serious. Therefore, in order to prevent traffic terror, the first thing is to implement the identity confirmation of passengers, followed by the through search of cargos of the passengers, supplement the capability against the terror by the security team members, strengthen the responsive capability of citizens, thorough check on facilities, expansion of social security network and others. All terrors are within the reach of our neighborhood at the time when we neglect it, and the most secured and safe way to prevent the terror is to check and prepare at all times beginning from our familiar facilities or means.

• Journal of Intelligence and Information Systems
• /
• v.12 no.2
• /
• pp.1-15
• /
• 2006

Currently the mobile web service is growing with a tremendous speed and mobile contents are spreading extensively. However, it is hard to search what the user wants because of some limitations of cellular phones. And the music is the most popular content, but many users experience frustrations to search their desired music. To solve these problems, this research proposes a hybrid recommendation system, MOBICORS-music (MOBIle COntents Recommender System for Music). Basically it follows the procedure of Collaborative Filtering (CF) system, but it uses Contents-Based (CB) data representation for neighborhood formation and recommendation of new music. Based on this data representation, MOBICORS-music solves the new item ramp-up problem and results better performance than existing CF systems. The procedure of MOBICORS-music is explained step by step with an illustrative example.