• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.05a
• /
• pp.559-562
• /
• 2011

In this research, We derived Recursive Least Squares(RLS) algorithm with adaptive maximum-likelihood channel estimate for digital pulse amplitude modulated sequence in the presence of intersymbol interference and additive white Gaussian noise. RLS algorithms have better convergence characteristics than conventional algorithms, LMS (Least Mean Squares) algorithms.

• Journal of information and communication convergence engineering
• /
• v.12 no.2
• /
• pp.128-134
• /
• 2014

In this paper, we present a graphics processing unit (GPU)-based matching technique for the purpose of fast feature matching between different images. The scale invariant feature transform algorithm developed by Lowe for various feature matching applications, such as stereo vision and object recognition, is computationally intensive. To address this problem, we propose a matching technique optimized for GPUs to perform computations in less time. We optimize GPUs for fast computation of keypoints to make our system quick and efficient. The proposed method uses a self-organizing map feature matching technique to perform efficient matching between the different images. The experiments are performed on various image sets to examine the performance of the system under varying conditions, such as image rotation, scaling, and blurring. The experimental results show that the proposed algorithm outperforms the existing feature matching methods, resulting in fast feature matching due to the optimization of the GPU.

• Journal of Institute of Control, Robotics and Systems
• /
• v.2 no.3
• /
• pp.148-157
• /
• 1996

Availability of input trajectories corresponding to desired output trajectories is often important in designing control systems for batch and other transient processes. In this paper, we propose a predictive control-type model-based iterative learning algorithm which is applicable to finding the nominal input trajectories of a linear time-invariant batch process. Unlike the other existing learning control algorithms, the proposed algorithm can be applied to nonsquare systems and has an ability to adjust noise sensitivity as well as convergence rate. A simple model identification technique with which performance of the proposed learning algorithm can be significantly enhanced is also proposed. Performance of the proposed learning algorithm is demonstrated through numerical simulations.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.41 no.3
• /
• pp.311-319
• /
• 1992

In this paper, a direct adaptive pole placement controller for an unknown linear time-invariant single-input single-output nonminimum phase plant is proposed. To design this direct adaptive pole placement controller, the auxiliary signals are introduced. Consequently, a linear equation error model is formulated for estimating both the controller parameters and the additional auxiliary parameters. To estimate the controller parameters and the additional auxiliary parameters, the exponentially weighted least-squares algorithm is implemented, and a method of selecting the characteristic polynomials of the sensitivity function filters is proposed. In this method, all the past measurement data are weighted exponentially. A series of simulations for a nonminimum phase plant is presented to illustrate some features of both the parameter estimation and the output response of this adaptive pole placement controller.

• Proceedings of the KIEE Conference
• /
• 1990.07a
• /
• pp.60-63
• /
• 1990

This note presents a direct adaptive pole assignment control for general discrete, linear, time-invariant, nonmimum phase system.Controller parameters are estimated from the recursive least-squares algorithm, and some additional auxiliary parameters are obtained from aset of recursive equations based on a certain polynomial identity which is derived from the pole assignment equation and the Bezout identity. This scheme increase the numerical stability of the auxiliary parameters, and guarantees local convergence without any extra conditions for the external input. The effectiveness of the proposed scheme is demonstrated by the computer simulation.

• Journal of information and communication convergence engineering
• /
• v.11 no.4
• /
• pp.298-306
• /
• 2013

The scale invariant feature transform (SIFT) is an effective algorithm used in object recognition, panorama stitching, and image matching. However, due to its complexity, real-time processing is difficult to achieve with current software approaches. The increasing availability of parallel computers makes parallelizing these tasks an attractive approach. This paper proposes a novel parallel approach for SIFT algorithm implementation using a block filtering technique in a Gaussian convolution process on the SIMD Pixel Processor. This implementation fully exposes the available parallelism of the SIFT algorithm process and exploits the processing and input/output capabilities of the processor, which results in a system that can perform real-time image and video compression. We apply this implementation to images and measure the effectiveness of such an approach. Experimental simulation results indicate that the proposed method is capable of real-time applications, and the result of our parallel approach is outstanding in terms of the processing performance.

• International journal of advanced smart convergence
• /
• v.4 no.2
• /
• pp.170-176
• /
• 2015

The stamp is one of the crucial information of traditional historical and cultural for nations. In this paper, we purpose to detect official stamps from scanned document and recognize the Mongolian traditional, historical stamps. Therefore we performed following steps: first, we detect official stamps from scanned document based on red-color segmentation and document standard. Then we collected 234 traditional stamp images with 6 classes and 100 official stamp images from scanned document images. Also we implemented the processing algorithms for noise removing, resize and reshape etc. Finally, we proposed a new scale invariant classification algorithm based on KNN (k-nearest neighbor). In the experimental result, our proposed a method had shown proper recognition rate.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.11
• /
• pp.1464-1471
• /
• 1999

This paper introduces complex-valued competitive learning neural network for spatio-temporal pattern recognition. There have been quite a few neural networks for spatio-temporal pattern recognition. Among them, recurrent neural network, TDNN, and avalanche model are acknowledged as standard neural network paradigms for spatio-temporal pattern recognition. Recurrent neural network has complicated learning rules and does not guarantee convergence to global minima. TDNN requires too many neurons, and can not be regarded to deal with spatio-temporal pattern basically. Grossberg's avalanche model is not able to distinguish long patterns, and has to be indicated which layer is to be used in learning. In order to remedy drawbacks of the above networks, unsupervised competitive learning using complex umber is proposed. Suggested neural network also features simultaneous recognition, time-shift invariant recognition, stable categorizing, and learning rate modulation. The network is evaluated by computer simulation with randomly generated patterns.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.40 no.4
• /
• pp.427-438
• /
• 1991

An iterative learning control scheme for exact-tracking control and parameter estimation of uncertain robotic system is preented. In the learning control structure, the control input converges globally and asymtotically to the desired input as iteration increases. Since convergence of parameter errors depends only on the persistent exciting condition of system trajectories along the iteration independently of the time-duration of trajectories, it may be achieved with system trajectories with small duration. In addition, the proposd learning control schemes are applicable to time-varying parametric systems as well as time-invariant systems, because the parameter estimation is performed at each fixed time along the iteration. In the parameter estimator, the acceleration information as well as the inversion of estimated inertia matrix are not used at all, which makes the proposed learning control schemes more feasible.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2000.11a
• /
• pp.443-448
• /
• 2000

In this paper, we propose a novel adaptive feedforward controller for periodic disturbance and noise cancellation, with a frequency tracking capability. It can be added to an existing feedback control system without altering the original closed-loop characteristics, which is based on adaptive algorithm. We introduce novel algorithm "Constrained AFC(adaptive feedforward controller) algorithm" that increase the convergence region regardless of the delay in the closed loop system. In the algorithms, coefficients of the controller are adapted using the residuals of constrained structure which are defined in such a way that the coefficients become time invariant. The proposed algorithm not only estimate the magnitude and phase of the tonal disturbance and noise but also track the frequency of the tone, which changes in quasi-static manner. The frequency tracking algorithm uses the instantaneous frequency approach based on Hilbert transform. A number of computer simulations have been carried out in order to demonstrate the effectiveness of proposed method under various conditions.