• 융합신호처리학회논문지
• /
• 제12권2호
• /
• pp.91-95
• /
• 2011

In this paper, we present an improved multi-scale gradient algorithm. The proposed algorithm works the effectively handling of both step and blurred edges. In the proposed algorithm, the image sharpening operator is sharpening the edges and contours of the objects. This operation gives an opportunity to get noise reduced image and step edged image. After that, multi-scale gradient operator works on noise reduced image in order to get a gradient image. The gradient image is segmented by watershed transform. The approach of region merging is used after watershed transform. The region merging is carried out according to the region area and region homogeneity. The region number of the proposed algorithm is 36% shorter than that of the existing algorithm because the proposed algorithm produces a few irrelevant regions. Moreover, the computational time of the proposed algorithm is relatively fast in comparison with the existing one.

• 융합신호처리학회 학술대회논문집
• /
• 한국신호처리시스템학회 2000년도 하계종합학술대회논문집
• /
• pp.309-312
• /
• 2000

In this work, a new algorithm for canceling MRI artifact due to translational motion in image plane is described. In the previous approach, the motions in the x direction and the y direction are estimated simultaneously. By analyzing their features, each x and y directional motion is canceled by different algorithms in two steps. First, it is noticed that the x directional motion corresponds to a shift of the x directional spectrum of the MRI signal. Next, the y directional motion is canceled by using a new constraint condition. This algorithm is shown to be effective by using a phantom image with simulated motion.

• 융합신호처리학회 학술대회논문집
• /
• 한국신호처리시스템학회 2000년도 하계종합학술대회논문집
• /
• pp.273-276
• /
• 2000

자동차 번호판은 조명과 카메라에 따라 영상에서 다양한 형태로 나타나고 영상내의 잡음으로 인해 알고리즘 방식으로 자동차 번호판을 인식하기가 쉽지 않다. 이러한 문제에 적합한 해결 방법으로 본 논문에서는 학습 기반의 자동차 번호판 인식 시스템을 제안한다. 제안한 시스템은 자동차 검출 모듈, 번호판 추출 모듈, 번호판 문자인식 모듈로 구성된다. 본 논문에서는 자동차 번호판 추출을 위해서 시간-지연 신경망(Time-Delay Neural Networks : TDNN)과 번호판 인식을 위해서 일반적인 신경망보다 일반화 성능이 뛰어난 서포트 벡터 머신(Support Vector Machines : SVMs)을 시스템에 적용한다. 주차장과 톨케이트에서 여러 시간대의 움직이는 자동차 영상들을 실험한 결과, 자동차 검출율은 100%, 번호판 추출율은 97.5%, 번호판 문자 인식율은 97.2%의 성능을 내었고, 전체 시스템 성능은 94.7%이며 처리 시간은 약 1초 미만이다. 따라서 본 논문에서 제안한 시스템은 실세계에서 유용하게 적용될 수 있다.

• 융합신호처리학회 학술대회논문집
• /
• 한국신호처리시스템학회 2000년도 하계종합학술대회논문집
• /
• pp.165-168
• /
• 2000

In this paper, the polyphase filter which has good ripple characteristic in the passband is proposed. This filter consists of the digital all-pass filter of parallel structure and it is the half-band filter with all zeros in unit circle. To approach easily in designing hardware, we determined the coefficients to the 16bit 1.15 format. To evaluate the performance of this filter, we analyzed the phase characteristic in each branch and simulated each filter with small coefficients. From the result, we have got to good ripple characteristic and also analyzed the fifth and the seventh, and compared them with four part : ripple, group delay, transition bandwidth, stopband attenuation.

• 한국산업융합학회 논문집
• /
• 제22권5호
• /
• pp.561-573
• /
• 2019

This study proposes a new approach to implement smart factory based on robot monitoring system for a small and medium sized enterprise. The simulator includes the forward kinematics and inverse kinematics analyzes of articulated robots. We also modeled the dynamics of the robot and made it possible to simulate it in the program. We studied the interface between the user and the monitoring programming system using the GUI environment of Windows OS, and it is configured to select the appropriate trajectory planning and control method. The reliability of simulator is illustrated by simulation test.

• 한국산업융합학회 논문집
• /
• 제21권6호
• /
• pp.369-378
• /
• 2018

This study proposes a new approach to real time implimemtation of shape recognition technology of die casting and forging parts based on robot vision for smart factory. The proposed shape recognition and inspection technology for forging and die casting parts is very useful for manufacturing process automatiom and smart factory including external form's automatic inspection of machanical or electronic panrs for the precision verification. The reliabiblity of proposed technology Ihas been illustrated through experiments.

• 한국산업융합학회 논문집
• /
• 제22권4호
• /
• pp.415-425
• /
• 2019

This study proposes a new approach to control Moving Stuff Action Through Iterative Learning robot with dual arm for smart factory. When robot moves object with dual arm, not only position of each hand but also contact force at surface of an object should be considered. However, it is not easy to determine every parameters for planning trajectory of the an object and grasping object concerning about variety compliant environment. On the other hand, human knows how to move an object gracefully by using eyes and feel of hands which means that robot could learn position and force from human demonstration so that robot can use learned task at variety case. This paper suggest a way how to learn dynamic equation which concern about both of position and path.

• 한국산업융합학회 논문집
• /
• 제21권5호
• /
• pp.191-195
• /
• 2018

By using an ANSYS product suite (CFX, Ansys Multiphysics), which is a powerful tool for multiphysics analysis of complicated physical phenomena, we performed a structural stress analysis based on fluid flow and heat transfer phenomena within a quick connect/disconnect (QC/DC) bellows system. Considering the extremely low temperatures in the QC/DC environment, an approach to the problem based on complex multi-physics phenomena, where different phenomena interact with each other, is crucial. Therefore, we use a numerical analysis technique where fluid-thermal-structural interactions are combined. In conclusion, when low temperature fluids flow inside bellows, the expected service life is conspicuously reduced due to the thermal stress caused by heat transfer. Therefore, in future research, a structure with considerably reduced thermal stress by robust design optimization will be derived.

• International journal of advanced smart convergence
• /
• 제8권1호
• /
• pp.75-81
• /
• 2019

Data imbalance problem is common and causes serious problem in machine learning process. Sampling is one of the effective methods for solving data imbalance problem. Over-sampling increases the number of instances, so when over-sampling is applied in imbalanced data, it is applied to minority instances. Under-sampling reduces instances, which usually is performed on majority data. We apply under-sampling and over-sampling to imbalanced data and generate sampled data sets. From the generated data sets from sampling and original data set, we construct a heterogeneous ensemble of classifiers. We apply five different algorithms to the heterogeneous ensemble. Experimental results on an intrusion detection dataset as an imbalanced datasets show that our approach shows effective results.

• 한국산업융합학회 논문집
• /
• 제23권6_1호
• /
• pp.889-896
• /
• 2020

We describe a new approach to the analyze the control performance of robotic manipulator based on the monitoring system. The structure of monitoring simulator is consist of seven modes such as control state mode, coordinate mode, input/output mode, program mode, parameters mode, and track mode. The applied control algorithme consists of an time varying feed-forward and feedback controller. The proposed scheme is simple in structure, fast in computation, and suitable for real-time implimemtation. Moreover, this scheme does not require any accurate dynamic modeling and values of parameters. Performance of the proposed monitoring system is illustrated by simulation and experiment for robot manipulator with six degrees of freedom.