• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.16 no.2
• /
• pp.104-110
• /
• 2016

Dense local image descriptors like SIFT are fruitful for capturing salient information about image, shown to be successful in various image-related tasks when formed in bag-of-words representation (i.e., histograms). In this paper we consider to utilize these dense local descriptors in the object tracking problem. A notable aspect of our tracker is that instead of adopting a point estimate for the target model, we account for uncertainty in data noise and model incompleteness by maintaining a distribution over plausible candidate models within the Bayesian framework. The target model is also updated adaptively by the principled Bayesian posterior inference, which admits a closed form within our Dirichlet prior modeling. With empirical evaluations on some video datasets, the proposed method is shown to yield more accurate tracking than baseline histogram-based trackers with the same types of features, often being superior to the appearance-based (visual) trackers.

• Journal of Korea Multimedia Society
• /
• v.15 no.4
• /
• pp.417-424
• /
• 2012

In this paper, we propose an automatic system that recognizes dynamic human gestures activity, including Arabic numbers from 0 to 9. We assume the gesture trajectory is almost in a plane that called principal gesture plane, then the Least Squares Method is used to estimate the plane and project the 3-D trajectory model onto the principal. An improved FNN model combined with HMM is proposed for dynamic gesture recognition, which combines ability of HMM model for temporal data modeling with that of fuzzy neural network. The proposed algorithm shows that satisfactory performance and high recognition rate.

• Proceedings of the KIEE Conference
• /
• 2009.07a
• /
• pp.1841_1842
• /
• 2009

본 논문에서는 WSN(Wireless Sensor Network)에서 RSSI(Receive Strength Signal Indicator)를 이용해 미지노드의 위치측정을 위한 퍼지 모델링 기법을 제안한다. RSSI는 거리에 따른 전파의 감쇠를 나타내는 것으로 측정 환경에 따라 신호 반사 및 잡음의 영향에 민감하다. 본 논문에서는 퍼지를 이용하여 측정된 RSSI를 거리로 환산하고 가장 짧은 거리와 그에 따른 거리오차를 모델링한다. 출력으로 입력 거리에 따른 가중치를 얻은 뒤 가중치를 적용한 거리의 무게 중심을 구하고 실제 미지노드의 위치와 비교함으로써 제안한 기법이 응용 가능함을 증명한다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 1998.11a
• /
• pp.202-207
• /
• 1998

본 논문에서는 유전자 알고리즘을 이용하여 불확실한 비선형 시스템의 퍼지-신경 회로망 모델링을 제안하였다. 제안한 퍼지-신경 회로망 모델링을 위한 학습 알고리즘은 다음과 같은 세 단계로 나누어 진행한다. 첫 번째 단계에서는 퍼지 모델의 소속 함수의 중심간과 표준편차를 구하여 초기 퍼지소속 함수를 결정한다. 두 번째 단계에서는 새로운 알고리즘을 통하여 언어적 퍼지 규칙을 만든다. 마지막 세 번째 단계에서는 유전자 알고리즘을 이용하여 중심값과 표준편차를 최적화함으로써 퍼지 모델의 소속 함수를 조절한다. 제안된 유전자 알고리즘의 장점은 흔히 신경 회로망에서 널리 쓰이는 역전파 알고리즘이 갖는 지역 최소점에 빠지는 현상이 없다는 것이다. 제안한 알고리즘의 유용성을 확인하기 위하여 일반적으로 가장 많이 쓰이는 비선형 시스템에 대하여 시뮬레이션 하여 확인하였다.

• Proceedings of the KIEE Conference
• /
• 2004.07d
• /
• pp.2513-2515
• /
• 2004

본 논문에서는 주어진 데이터를 나누어 부분공간으로 구성하여 특성을 구분하거나 또다른 모델의 입력 파라미터로 제공하는 방법 중 하나의 클러스터링의 성능 개선과 이를 이용하여 퍼지 모델링을 실시하였다. 일반적인 클러스터링에서 볼 수 있는 초기 파라미터 결정 문제와 알고리즘의 수렴 문제에 대하여 문제점을 개선하였으며 클러스터링에 의하여 추정된 파라미터를 퍼지 모델에 적용하였다. 또한 일반적인 퍼지 모델의 경우 각 입력의 차원이 서로 독립적으로 구성되어 있어 데이터에서 존재하는 입력간의 상관관계를 고려하지 않았다. 제안된 퍼지 모델에서는 클러스터링에서 추정된 입력간의 상관관계(공분산)까지 고려하여 모델의 성능을 개선하였다. 제안된 논문의 유용성을 시뮬레이션을 통하여 보이고자 한다.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.34C no.9
• /
• pp.61-67
• /
• 1997

The HMM-Net is an architecture for a neural network that implements a hidden markov model(HMM). The architecture is developed for the purpose of combining the classification power of neural networks with the time-domain modeling capability of HMMs. Criteria which are used for learning HMM_Net classifiers are maximum likelihood(ML), maximum mutual information (MMI), and minimization of mean squared error(MMSE). In this classifiers trained by the gradient descent algorithm with the above criteria. Experimental results for the isolated numbers from /young/to/koo/ show that in the binary inputs the performance of MMSE is better than the others, while in the fuzzy inputs the performance of MMI is better than the others.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2003.04a
• /
• pp.282-287
• /
• 2003

DTTS(Dynamic Track Tensioning System) system requires robust control performance for the various maneuvering tasks. However, it is very difficult to tune the controller gains in experiments. In this paper, the hydraulic unit is modeled and constructed into the DTTS control module in Matlab/Simulink The control module is interfaced to the vehicle dynamics module so that the control performance of the DTTS system can be evaluated in simulations. The dynamics data and control input data are exchanged between two modules at each control time-step. The gains in the fuzzy-logic controller are varied and the control performance is evaluated in simulations. The proposed simulation tool can be very useful for the gain tuning of track tension controller in bucked vehicles

• Journal of Institute of Control, Robotics and Systems
• /
• v.13 no.2
• /
• pp.173-178
• /
• 2007

This paper is to establish an appropriate model for predicting the temperature decreases in the batch transferred from the refining process to the caster in steel-making companies. Mathematical modeling of the temperature decreases between the processes is difficult, since the reaction mechanism by which the temperature changes in a molten steel batch is dynamic, uncertain and complex. Three soft computing techniques are examined using the same data, namely the multiple regression, fuzzy regression, and neural net (NN) models. To compare the accuracy of these three models, a limited number of input variables are selected from those variables significantly affecting the temperature decrease. The results show that the difference in accuracy between the three models is not statistically significant. Nonetheless, the NN model is recommended because of its adaptive ability and robustness. The method presented in this paper allows the temperature decrease to be predicted without requiring any precise metallurgical knowledge.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.3 no.2
• /
• pp.161-165
• /
• 2003

In this paper, we investigate the issues on the design and implementation of tele-operation system based on the haptic interface. Here, the 3-DOF haptic device and the X-Y-Z stage are employed as master controller and slave system respectively. For this master-slave system, the force feedback algorithm, the modeling of virtual environments and the control method of X-Y-Z stage are presented. In this paper, internet network is used for data communication between master and slave. We construct virtual environment of the real convex surface from the force-feedback in controlling the X-Y-Z stage and measuring the force applied by the 3-DOF haptic device.

• Journal of the Korean Society for Precision Engineering
• /
• v.20 no.12
• /
• pp.198-204
• /
• 2003

The bone fracture healing is simulated by using one of the complex system rules, named cellular automata method. It is assumed that each cell has property of Bone, Cartilage or Fibrous connective tissue. Nine local rules are adopted to change the property of each cell against the mechanical stimulus, which consists of the strain energy density, and the existence of bone in the surroundings. Two dimensional sheep metatarsal model is considered and the bone fracture healing is simulated. The simulation results agree well with those obtained by using fuzzy logic model and experimental data. The cellular automata method found to be one of the simulation methods to express the bone fracture healing. The cellular automata method is expected to be effective in representing biological phenomenon.