• 한국군사과학기술학회지
• /
• 제13권5호
• /
• pp.824-830
• /
• 2010

It is not easy job to find a underwater target using sonar system in the ASW operations. Many researchers have tried to solve anti-submarine search problem aiming to maximize the probability of detection under limited searching conditions. The classical 'Search Theory' deals with search allocation problem and search path problem. In both problems, the main issue is to prioritize the searching cells in a searching area. The number of possible searching path that is combination of the consecutive searching cells increases rapidly by exponential function in the case that the number of searching cells or searchers increases. The more searching path we consider, the longer time we calculate. In this study, an effective algorithm that can maximize the probability of detection in shorter computation time is presented. We show the presented algorithm is quicker method than previous algorithms to solve search problem through the comparison of the CPU computation time.

• 한국IT서비스학회지
• /
• 제11권3호
• /
• pp.103-127
• /
• 2012

With the advent of SNS (Social Network Services), the product reviews by friends in SNS are intensively utilized for online marketing. However, there is a lack of empirical evidence on the actual marketing effect of SNS reviews, although we need to identify who can be the target of SNS marketing in terms of customer attributes, preferences, or experiences. In this study, we investigate the moderating role of customer attributes in identifying the effect of SNS reviews on customer purchasing decision. As the moderating variables, we adopt 'information search experience' and 'perception of information overload'. Research results evidence that, in order to understand the effect of SNS reviews in a comprehensive manner, we need to examine it in the context of various related factors such as 'information search experience' and 'perception of information overload'. The results show that the persuading effect of SNS reviews for product purchasing is stronger for the customers with the lower information search experiences as well as the lower perception on the information overload on the web. This result delivers managerial implications on who can be the target customers of SNS marketing.

• 한국지능시스템학회논문지
• /
• 제17권3호
• /
• pp.368-373
• /
• 2007

최근에 지능형 로봇분야에서 주위 카메라를 기반으로 실시간으로 환경인식 및 물체 추적 등 다양한 분야에서 연구가 활발히 진행되고 있다. 환경인식 및 물체 추적은 결국 배경과 관심물체를 분리하는 것이라고 볼 수 있는 데, 차 연산을 이용하여 물체의 움직임만을 배경으로 분리하는 방법과 물체인식을 통해 배경으로부터 분리하여 추적하는 방법에 대한 연구가 지속적으로 이루어지고 있다. 본 논문에서는 배경과 물체 사이에서 변화하는 색상의 변화를 퍼지기법을 이용하여 물체를 배경과 분리하여 실시간으로 물체를 추적하고자 한다. 실시간 물체 추적을 위해 전체영상에 대한 전역적 탐색을 통해 여러 후보 물체 중 관심물체를 배경에서 추출 후, 추출된 물체의 크기에 따른 지역탐색을 통하여 물체를 추적하는 방법이다. 그리고 본 논문에서는 ARM 프로세서를 이용한 카메라시스템을 제작하여 실시간으로 영상분활을 실험하였다.

• 대한인간공학회:학술대회논문집
• /
• 대한인간공학회 1993년도 추계학술대회논문집
• /
• pp.124-132
• /
• 1993

This study compared two prediction methods-regression and artificial neural network (ANN) on the visual search performance when monitoring a multi-parameter screen with different occurrence frequencies. Under the highlighting condition for the highest occurrence frequency parameter as a search cue, it was found from the requression analysis that variations of mean search time (MST) could be expained almost by three factors such as the number of parameters, the target occurrence frequency of a highlighted parameter, and the highlighted parameter size. In this study, prediction performance of ANN was evaluated as an alternative to regression method. Backpropagation method which was commonly used as a pattern associator was employed to learn a search behavior of subjects. For the case of increased number of parameters and incresed target occurrence frequency of a highlighted parameter, ANN predicted MST's moreaccurately than the regression method (p<0.000). Only the MST's predicted by ANN did not statistically differ from the true MST's. For the case of increased highlighted parameter size. both methods failed to predict MST's accurately, but the differences from the true MST were smaller when predicted by ANN than by regression model (p=0.0005). This study shows that ANN is a good predictor of a visual search performance and can substitute the regression method under certain circumstances.

• 한국군사과학기술학회지
• /
• 제15권5호
• /
• pp.586-593
• /
• 2012

The Navy is in the process of developing a sonar-operation strategy to increase the effectiveness of underwater target searching capability. HMS is the basic strategy to detect underwater targets. The advantages of HMS is that, it has a short preparation time to operate and can be always used regardless of sea conditions and weather. However, it is difficult to effectively detect underwater targets due to the interaction between marine environments and sonar-operations. During the research, the effectiveness of the HMS system's underwater target searching capability was analyzed by integrating various search and defense patterns, and environment conditions into the simulation. In the simulation the ship search an evasive target within a set region. The simulation presented results for an effective searching and defense methods of underwater targets. These research results can be used as foundation for advancing and improving the sonar operational tactics.

• 대한전자공학회논문지
• /
• 제25권1호
• /
• pp.63-71
• /
• 1988

In this paper, a fast matching method using hierarchical neighborhood search and subtemplate to reduce very heavy computational load of the conventional matching method, is presented. Some parameters of the proposed method are chosen so that an automatic target tracker to which it is applied can track one moving object well in comparatively simple background. Experimental results show that its performance is not so degraded in spite of high computational reduction over that of the matching method using 3-step search.

• 한국시뮬레이션학회논문지
• /
• 제28권1호
• /
• pp.23-39
• /
• 2019

군 병력 감축, 인명중시 사상의 확산 그리고 4차 산업혁명을 통한 혁신 기술의 국방에서의 활용과 같은 시대적 요구 속에 드론-봇의 적극적인 활용이 요구되고 있다. 특히, 드론을 활용한 감시 및 정찰은 미래 전장에서 큰 역할을 할 것으로 예상된다. 하지만 정찰용 드론의 운용 개념, 특히 적을 찾기 위한 드론의 탐색 경로에 대한 연구는 많이 이루어지지 않았다. 본 연구에서는 정찰 목적용 드론의 효과적인 운용을 위한 탐색 방법을 시뮬레이션 분석을 통해 제안한다. 시뮬레이션에서 드론과 적은 연속공간에서 선형(First-Order)으로 움직이며, 적은 불확실성을 반영하여 랜덤워크 기법을 적용하였다. 연구는 먼저 기존에 군에서 주요하게 활용하던 탐색 방법(Parallel, Spiral)이 실제로 목표를 탐지하는 확률을 제시하며, 이어서 탐지자의 탐색반경과 속도가 탐지 확률에 미치는 영향을 분석한다. 마지막으로, 적이 랜덤하게 이동하지 않고 특정한 목표를 가지고 이동할 때 적용할 수 있는 새로운 탐지방법인 PS(Probability Search), PCS((Probability Circle Search), HS(Hamiltonian Search), HCS(Hamiltonian Circle Search) 방법을 소개하고 이에 대한 실험결과를 제시한다. 본 연구에서 제시한 탐색방법은 드론의 정찰 작전 시 활용도가 클 것으로 기대한다.

• 산업경영시스템학회지
• /
• 제40권4호
• /
• pp.260-267
• /
• 2017

Quantum-inspired Genetic Algorithm (QGA) is a probabilistic search optimization method combined quantum computation and genetic algorithm. In QGA, the chromosomes are encoded by qubits and are updated by quantum rotation gates, which can achieve a genetic search. Asset-based weapon target assignment (WTA) problem can be described as an optimization problem in which the defenders assign the weapons to hostile targets in order to maximize the value of a group of surviving assets threatened by the targets. It has already been proven that the WTA problem is NP-complete. In this study, we propose a QGA and a hybrid-QGA to solve an asset-based WTA problem. In the proposed QGA, a set of probabilistic superposition of qubits are coded and collapsed into a target number. Q-gate updating strategy is also used for search guidance. The hybrid-QGA is generated by incorporating both the random search capability of QGA and the evolution capability of genetic algorithm (GA). To observe the performance of each algorithm, we construct three synthetic WTA problems and check how each algorithm works on them. Simulation results show that all of the algorithm have good quality of solutions. Since the difference among mean resulting value is within 2%, we run the nonparametric pairwise Wilcoxon rank sum test for testing the equality of the means among the results. The Wilcoxon test reveals that GA has better quality than the others. In contrast, the simulation results indicate that hybrid-QGA and QGA is much faster than GA for the production of the same number of generations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권12호
• /
• pp.3976-3990
• /
• 2022

Currently, cross-attention is widely used in Siamese trackers to replace traditional correlation operations for feature fusion between template and search region. The former can establish a similar relationship between the target and the search region better than the latter for robust visual object tracking. But existing trackers using cross-attention only focus on rich semantic information of high-level features, while ignoring the appearance information contained in low-level features, which makes trackers vulnerable to interference from similar objects. In this paper, we propose a Multi-level Cross-attention Siamese network(MCSiam) to aggregate the semantic information and appearance information at the same time. Specifically, a multi-level cross-attention module is designed to fuse the multi-layer features extracted from the backbone, which integrate different levels of the template and search region features, so that the rich appearance information and semantic information can be used to carry out the tracking task simultaneously. In addition, before cross-attention, a target-aware module is introduced to enhance the target feature and alleviate interference, which makes the multi-level cross-attention module more efficient to fuse the information of the target and the search region. We test the MCSiam on four tracking benchmarks and the result show that the proposed tracker achieves comparable performance to the state-of-the-art trackers.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제6권3호
• /
• pp.264-269
• /
• 2006

This paper presents the strategy for searching a hidden object in an unknown area for using by multiple distributed autonomous robotic systems (DARS). To search the target in Markovian space, DARS should recognize th ε ir surrounding at where they are located and generate some rules to act upon by themselves. First of all, DARS obtain 6-distances from itself to environment by infrared sensor which are hexagonally allocated around itself. Second, it calculates 6-areas with those distances then take an action, i.e., turn and move toward where the widest space will be guaranteed. After the action is taken, the value of Q will be updated by relative formula at the state. We set up an experimental environment with five small mobile robots, obstacles, and a target object, and tried to research for a target object while navigating in a un known hallway where some obstacles were placed. In the end of this paper, we present the results of three algorithms - a random search, an area-based action making process to determine the next action of the robot and hexagon-based Q-learning to enhance the area-based action making process.