• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.4 no.5
• /
• pp.859-876
• /
• 2010

In this paper, we propose a hierarchical application traffic classification system as an alternative means to overcome the limitations of the port number and payload based methodologies, which are traditionally considered traffic classification methods. The proposed system is a new classification model that hierarchically combines a binary classifier SVM and Support Vector Data Descriptions (SVDDs). The proposed system selects an optimal attribute subset from the bi-directional traffic flows generated by our traffic analysis system (KU-MON) that enables real-time collection and analysis of campus traffic. The system is composed of three layers: The first layer is a binary classifier SVM that performs rapid classification between P2P and non-P2P traffic. The second layer classifies P2P traffic into file-sharing, messenger and TV, based on three SVDDs. The third layer performs specialized classification of all individual application traffic types. Since the proposed system enables both coarse- and fine-grained classification, it can guarantee efficient resource management, such as a stable network environment, seamless bandwidth guarantee and appropriate QoS. Moreover, even when a new application emerges, it can be easily adapted for incremental updating and scaling. Only additional training for the new part of the application traffic is needed instead of retraining the entire system. The performance of the proposed system is validated via experiments which confirm that its recall and precision measures are satisfactory.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.39 no.5
• /
• pp.45-55
• /
• 2002

In this paper, we describe an outdoor augmented reality system using GPS tracking for position and attitude information. The system consist of a remote mobile operation unit and a ground operation unit. The remote mobile operation unit includes a real-time image acquiring device, a GPS tracking device, and a wireless data transceiver; the ground operation unit includes a wireless transceiver, a virtual image generating device, and an image superimposing device. The GPS tracking device for measurement of position and attitude of the remote mobile operation unit was designed by TANS Vector and RT-20 for DGPS. The wireless data transceiver was for data transmission between the remote mobile operation unit and the ground operation unit. After the remote mobile operation unit was installed on a vehicle and a helicopter, the system was evaluated to verify its validity in actual applications. It was found that the implemented system could be used for obtaining real-time remote information such as construction simulation, tour guide, broadcasting, disaster observation, or military purpose.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.12
• /
• pp.2946-2952
• /
• 2014

Among the Example based Super Resolution(SR) techniques, Neighbor embedding(NE) has been inspired by manifold learning method, particularly locally linear embedding. However, the poor generalization of NE decreases the performance of such algorithm. The sizes of local training sets are always too small to improve the performance of NE. We propose the Learning Sparse-Neighbor Image Representation baesd on SVR having an excellent generalization ability to solve this problem. Given a low resolution image, we first use bicubic interpolation to synthesize its high resolution version. We extract the patches from this synthesized image and determine whether each patch corresponds to regions with high or low spatial frequencies. After the weight of each patch is obtained by our method, we used to learn separate SVR models. Finally, we update the pixel values using the previously learned SVRs. Through experimental results, we quantitatively and qualitatively confirm the improved results of the proposed algorithm when comparing with conventional interpolation methods and NE.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.6
• /
• pp.19-26
• /
• 2014

An ensemble classifier system is a widely-used multi-classifier system, which combines the results from each classifier and, as a result, achieves better classification result than any single classifier used. Several methods have been used to build an ensemble classifier including boosting, which is a cascade method where misclassified examples in previous stage are used to boost the performance in current stage. Boosting is, however, a serial method which does not form a complete feedback loop. In this paper, proposed is context sensitive SVM ensemble (CASE) which adopts SVM, one of the best classifiers in term of classification rate, as a basic classifier and clustering method to divide feature space into contexts. As CASE divides feature space and trains SVMs simultaneously, the result from one component can be applied to the other and CASE achieves better result than boosting. Experimental results prove the usefulness of the proposed method.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.4 s.304
• /
• pp.119-126
• /
• 2005

We propose a new content-based image retrieval using a representative color histogram and directional pattern histogram that is adaptive to the classification characteristics of the image blocks. In the proposed method the color and pattern feature vectors are extracted according to the characteristics o: the block classification after dividing the image into blocks with a fixed size. First, the divided blocks are classified as either luminance or color blocks depending on the saturation of the block. Thereafter, the color feature vectors are extracted by calculating histograms of the block average luminance co-occurrence for the luminance block and the block average colors for the color blocks. In addition, block directional pattern feature vectors are extracted by calculating histograms after performing the directional gradient classification of the luminance. Experimental results show that the proposed method can outperform the conventional methods as regards the precision and the size of the feature vector dimension.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.6
• /
• pp.81-89
• /
• 2013

Currently Internet shopping (or shopping online) is becoming the common consumption channel for Chinese, and it is more likely to continue to grow. Although E-tailers (or the Internet shopping mall) in China is rapidly growing, there are not very many shopping malls that can meet customer satisfaction. E-tailers in Korea analyze the quality evaluation and customer satisfaction of shopping malls. If the Internet shopping that is suitable for Chinese students studying in Korea is built, it is expected to strengthen international competitive power. In this paper, the comparative analysis of Customer satisfaction for Internet shopping between Chinese students studying in Korea and Korean university students is provided. Furthermore, we analyze the customer satisfaction model of Chinese students studying in Korea by using the multi lease square support vector machine that obtains the global optimal solution. Analysis of customer satisfaction of Chinese students studying in Korea are not only used for E-tailers in Korea, but it can strengthen international competitive power.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.18 no.1 s.116
• /
• pp.53-61
• /
• 2007

This paper researches on the scheme superimposing the rotation phases over the pilot and data symbols in order to reduce the peak-to-average power ratio(PAPR) of the orthogonal frequency division multiplexing(OFDM) communication. The bandwidth and power efficiency are the main consideration. The phases of rotation vector are added to those of both pilot symbols and data symbols interlaying between any two pilot symbols in an OFDM block. Owing to this scheme the transmitter reduces the PAPR using the partial transmit sequences(PTS) and the receiver restores the data symbol utilizing the channel estimation of pilot symbols. Therefore, the bandwidth efficiency is accomplished by not using the further subcarriers for the reduction of PAPR and the enormous increase of bit error rate according to the receiving error of the side information, i.e. the phases of rotation vector, is prevented. In other words, both bandwidth-and power-efficiency and quality of communication performance can be improved.

• Journal of Broadcast Engineering
• /
• v.17 no.1
• /
• pp.17-25
• /
• 2012

In this paper, we propose a method of producing a stereoscopic image from multiple images captured from a monoscopic camera. By translating a camera in the horizontal direction, left and right images are chosen among N captured images. For this, image edges are extracted and a rotational angle is estimated from edge orientation. Also, a translational vector is also estimated from the correlation of projected image data. Then, two optimal images are chosen and subsequently compensated using the rotational angle as well as the translational vector in order to make a satisfactory stereoscopic image. The proposed method was performed on thirty-two test image set. The subjective visual fatigue test was carried out to validate the 3D quality of stereoscopic images. In terms of visual fatigue, the 3D satisfaction ratio reached approximately 84%.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.11
• /
• pp.31-38
• /
• 2013

Feature normalization as a pre-processing step has been widely used in classification problems to reduce the effect of different scale in each feature dimension and error as a result. Most of the existing methods, however, assume some distribution function on feature distribution. Even worse, existing methods do not use the labels of data points and, as a result, do not guarantee the optimality of the normalization results in classification. In this paper, proposed is a supervised rank normalization which combines rank normalization and a supervised learning technique. The proposed method does not assume any feature distribution like rank normalization and uses class labels of nearest neighbors in classification to reduce error. SVM, in particular, tries to draw a decision boundary in the middle of class overlapping zone, the reduction of data density in that area helps SVM to find a decision boundary reducing generalized error. All the things mentioned above can be verified through experimental results.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.731-734
• /
• 2009

In this paper, we propose the method the vehicle color recognizing in the image including a vehicle. In an image, the color feature vector of a vehicle is extracted and by using the backpropagation learning algorithm, that is the multi-layer perceptron, the recognized vehicle color. By using the RGB and HSI color model the feature vector used as the input of the backpropagation learning algorithm is the feature of the color used as the input of the neural network. The color of a vehicle recognizes as the white, the silver color, the black, the red, the yellow, the blue, and the green among the color of the vehicle most very much found out as 7 colors. By using the image including a vehicle for the performance evaluation of the method proposing, the color recognition performance was experimented.