• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.9C
• /
• pp.861-866
• /
• 2007

A clustering algorithm for Gaussian Probability Distribution Function (GPDF) data called Centroid Neural Network with a Bhattacharyya Kernel (BK-CNN) is proposed in this paper. The proposed BK-CNN is based on the unsupervised competitive Centroid Neural Network (CNN) and employs a kernel method for data projection. The kernel method adopted in the proposed BK-CNN is used to project data from the low dimensional input feature space into higher dimensional feature space so as the nonlinear problems associated with input space can be solved linearly in the feature space. In order to cluster the GPDF data, the Bhattacharyya kernel is used to measure the distance between two probability distributions for data projection. With the incorporation of the kernel method, the proposed BK-CNN is capable of dealing with nonlinear separation boundaries and can successfully allocate more code vector in the region that GPDF data are densely distributed. When applied to GPDF data in an image classification probleml, the experiment results show that the proposed BK-CNN algorithm gives 1.7%-4.3% improvements in average classification accuracy over other conventional algorithm such as k-means, Self-Organizing Map (SOM) and CNN algorithms with a Bhattacharyya distance, classed as Bk-Means, B-SOM, B-CNN algorithms.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.12C
• /
• pp.1044-1051
• /
• 2010

In this paper, we propose an efficient soft-output signal detection method for spatially multiplexed multiple input multiple output (MIMO) systems. The proposed method is based on the ordered successive interference cancellation (OSIC) algorithm, but it significantly improves the performance of the original OSIC algorithm by solving the error propagation problem. The proposed method combines this enhanced OSIC (ESIC) algorithm with a multiple ordering technique in a very efficient way. As a result, the log likelihood ratio (LLR) values can be computed by using a very small set of candidate symbol vectors. The proposed method has been implemented with a $0.13{\mu}m$ CMOS technology for a $4{\times}4$ 16-QAM MIMO system. The simulation and implementation results show that the proposed detector provides a very good solution in terms of performance and hardware complexity.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.10
• /
• pp.55-62
• /
• 2014

In the intelligent video surveillance system, moving objects generally are detected by calculating difference between background and input image. However formation of reliable background is known to be still challenging task because it is hard to cope with the complicated background. In this paper we propose an adaptive background formation algorithm considering stationary object. At first, the initial background is formed by averaging the initial N frames. Object detection is performed by comparing the current input image and background. If the object is at a stop for a long time, we consider the object as stationary object and background is replaced with the stationary object. On the other hand, if the object is a moving object, the pixels in the object are not reflected for background modification. Because the proposed algorithm considers gradual illuminance change, slow moving object and stationary object, we can form background adaptively and robustly which has been shown by experimental results.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.1
• /
• pp.80-87
• /
• 2017

This paper proposes a pre-processing algorithm to improve perceptual video coding efficiency which decomposes an input frame via a sub-band decomposition, and suppresses only high frequency band(s) having low visual sensitivity. First, we decompose the input frame into several frequency subbands by a linear sub-band decomposition. Next, high frequency subband(s) which is rarely recognized by human visual system (HVS) is suppressed by applying relatively small gain(s). Finally, the high frequency suppressed frame is compressed by a specific video encoder. We can find from the experimental results that if comparing before-use and after-use of the proposed pre-processing prior to the encoder, no visual difference is shown. Also, the proposed algorithm achieves bit-saving of 13.12% on average in a H.264 video encoder.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.23 no.3
• /
• pp.395-404
• /
• 2013

A pattern matching algorithm plays an important role in traffic identification and classification based on predefined patterns for intrusion detection and prevention. As attacks become prevalent and complex, current patterns are written using regular expressions, called regexes, which are expressed into the deterministic finite automata(DFA) due to the guaranteed worst-case performance in pattern matching process. Currently, because of the increased complexity of regex patterns and their large number, memory-efficient DFA from states reduction have become the mainstay of pattern matching process. However, most of the previous works have focused on reducing only the number of states on a single automaton, and thus there still exists a state blowup problem under the large number of patterns. To solve the above problem, we propose a new state compression algorithm that merges states on multiple automata. We show that by merging states with the same input character on multiple automata, the proposed algorithm can lead to a significant reduction of the number of states in the original DFA by as much as 40.0% on average.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.3
• /
• pp.630-638
• /
• 2016

In this paper, a feature based panorama image generation algorithm using FAST(Features from Accelerated Segment Test) method that is faster than SIFT(Scale Invariant Feature Transform) and SURF(Speeded Up Robust Features) is proposed. Cylindrical projection is performed to generate natural panorama images with numerous images as input. The occurred error can be minimized by applying RANSAC(Random Sample Consensus) for the matching process. When we synthesize numerous images acquired from different camera angles, we use blending techniques to compensate the distortions by the heterogeneity of border line. In that way, we could get more natural synthesized panorama image. The proposed algorithm can generate natural panorama images regardless the order of input images and tilted images. In addition, the image matching can be faster than the conventional method. As a result of the experiments, distortion was corrected and natural panorama image was generated.

• Journal of Korea Multimedia Society
• /
• v.3 no.1
• /
• pp.92-100
• /
• 2000

For the operation of the practical software systems, the development of new software, extension and modification phases of current software are successively needed through the new requirements added and their errors debugging detected. We need to understand current program exactly during a working serial jobs. When we'd like to extract the statements which influence the variable of specific statement of program, we generally analyze the program behavior through execution trace of program for the input values. It is important to compute dynamic program slice related to all statements that actually affect the value of a variable occurrence for a given program input and to develop techniques of its implementation. But traditional dynamic slicing techniques are inconvenient to make program execution history at the beginning implementation. In this paper, we propose a new improved algorithm which can produce program slice by use of dynamic control information and program slicing techniques except execution history file. Also we can find that the proposed program slicing approach is more efficient than the traditional program slicing algorithm on the practical testing environment.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.7
• /
• pp.821-829
• /
• 2004

A face recognition has been used for verifying and authorizing valid users, but its applications have been restricted according to lighting conditions. In order to minimizing the restricted conditions, this paper proposes a new algorithm of detecting the face from the input image obtained under the irregular lighting condition. First, the proposed algorithm extracts an edge difference image from the input image where a skin color and a face contour are disappeared due to the background color or the lighting direction. In the next step, it extracts a face region using the histogram of the edge difference image and the intensity information. Using the intensity information, the face region is divided into the horizontal regions with feasible facial features. The each of horizontal regions is classified as three groups with the facial features(including eye, nose, and mouth) and the facial features are extracted using empirical properties of the facial features. Only when the facial features satisfy their topological rules, the face region is considered as a face. It has been proved by the experiments that the proposed algorithm can detect faces even when the large portion of face contour is lost due to the inadequate lighting condition or the image background color is similar to the skin color.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.9C
• /
• pp.729-734
• /
• 2010

In the initial stage of the communications between a base station and a satellite transponder, the base station transmits the frequency-sweeping un-modulated up-link carrier within a certain frequency range to acquire the doppler frequency shift and signal power between the base station and the satellite in orbital flight. The satellite transponder acquires and tracks the carrier in order to initialize the communication. To control such initialization process, the satellite receiver should analyze the input carrier signal in various ways. This paper presents an SNR estimation algorithm to control the initialization process. The proposed algorithm converts the input signal into the baseband polar coordinate representation and estimates the SNR via the statistics of the angular signal components as well as the status parameters to control the receiver. The Monte-Carlo simulations shows the validity of the estimation proposed.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.9A
• /
• pp.973-982
• /
• 2004

High-speed scheduling algorithms are required for high-performance input-queued switches to achieve good performance. Various Round-Robin scheduling algonthms for Virtual-Output-Queue (VOQ) switch architectures have been proposed, hke iSLIP, DRRM (Dual Round-Robin Matching). iSLIP can achieve high performance and have already been implemented in hardware. DRRM has been proved to achieve better performance and simpler than iSLIP But neither iSLIP nor DRRM can efficiently solve the problem of the Round-Robm pointers' desynchronization. In this paper, we have proposed "Well-Regulated Pseudo-request Dual Round-Robin Matching" Algorithm. It is developed from DRRM, and can always keep the pointers' desynchronization. Since our algorithm is based on the Round-Robin scheduling, it is also simple to be implemented. And simulation results also show that our proposed algonthm performs pretty well under various, traffic models.