• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.18 no.2
• /
• pp.179-189
• /
• 2000

This study is tried to organic combination in implementing the suitability analysis of urban landuse between GSIS and ANN(Artificial Neural Network). ANN has merit that can decide rationally connectivity weights among neural network nodes through procedure of learning. It is estimated to be possible that replacing the weight among factors needed in spatial analysis of the connectivity weight on neural network. This study is composed of two kinds of neural networks to be executed. First neural network was used in the suitability analysis of landuse and second one was oriented to analyze of optimum landuse pattern. These neural networks were learned with back-propagation algorithm using the steepest gradient which is embodied by C++ program and used sigmoid function as a active function. Analysis results show landuse suitability map and optimum landuse pattern of study area consisted of residental, commercial. industrial and green zone in present zoning system. Each result map was written by the Grid format of Arc/Info. Also, suitability area presented in the suitability map and optimum landuse pattern show distribution pattern consistent with theroretical concept or urban landuse plan in aspect of location and space structure.

• Journal of Biomedical Engineering Research
• /
• v.18 no.3
• /
• pp.223-231
• /
• 1997

In this paper, an interframe coding method for volumetric medical images is proposed. By treating interslice variations as the motion of bones or tissues, we use the motion compensation (MC) technique to predict the current frame from the previous frame. Instead of a block matching algorithm (BMA), which is the most common motion estimation (ME) algorithm in video coding, image warping with biolinear transformation has been suggested to predict complex interslice object variation in medical images. When an object disappears between slices, however, warping prediction has poor performance. In order to overcome this drawback, an overlapped block motion compensation (OBMC) technique is combined with carping prediction. Motion compensated residual images are then encoded by using an embedded zerotree wavelet (EZW) coder with small modification for consistent quality of reconstructed images. The experimental results show that the interframe coding suing warping prediction provides better performance compared with interframe coding, and the OBMC scheme gives some additional improvement over the warping-only MC method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.3
• /
• pp.458-468
• /
• 2009

The patching technique has been used for removing initial waiting time of NVOD. The technique was applicable only to time segmented NVOD services. However, data segmented NVOD methods have been known to have less initial waiting time or use less server's bandwidths than the time segmented NVODs. Therefore, we expect that if patching is applied to a data segmented NVOD, a more efficient NVOD service will be possible. But there has not been such approach. In this paper, we propose a new VOD scheduling algorithm that combines the patching technique with a data segmented NVOD, thus it uses less server's bandwidths and removes initial waiting time, hence makes a true VOD service possible. The proposed technique also absorbs the storage overhead necessary for client buffering incurred by patching into the data segmentation policy of the NVOD scheduling, hence does not need any extra overhead of the client's resource. Therefore, it is also applicable to the case that the client has resource limits.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39A no.5
• /
• pp.280-292
• /
• 2014

Recently, heterogeneous networks consisting of small-cells on top of traditional macro-cellular network has attracted much attention, because traditional macro-cellular network is not suitable to support more demanding mobile data traffic due to its limitation of spatial reuse. However, due to the transmit power difference between macro- and small-cells, most users are associated with macro-cells rather than small-cells. To solve this problem, enhanced inter-cell interference coordination (eICIC) has been introduced. Particularly, in eICIC, the small-cell coverage is forcibly expanded to associate more users with small-cells. Then, to avoid cross-tier interference from macro-cells, these users are allowed to receive the data during almost blank subframe (ABS) in which macro-cells almost remain silent. However, this approach is not sufficient to balance the load between macro- and small-cells because it only expands the small-cell coverage. In this paper, we propose a load balance scheme improving proportional fairness for heterogeneous networks employing eICIC. In particular, the proposed scheme combines the greedy-based user association and the ABS rate determination in a recursive manner to perform the load balance.

• Journal of Korea Multimedia Society
• /
• v.6 no.2
• /
• pp.216-223
• /
• 2003

In this paper, we present SCCRS(Speech and Character Combined Recognition System) for speaker /writer independent. on-line multimodal interfaces. In general, it has been known that the CHMM(Continuous Hidden Markov Mode] ) is very useful method for speech recognition and on-line character recognition, respectively. In the proposed method, the same CHMM is applied to both speech and character recognition, so as to construct a combined system. For such a purpose, 115 CHMM having 3 states and 9 transitions are constructed using MLE(Maximum Likelihood Estimation) algorithm. Different features are extracted for speech and character recognition: MFCC(Mel Frequency Cepstrum Coefficient) Is used for speech in the preprocessing, while position parameter is utilized for cursive character At recognition step, the proposed SCCRS employs OPDP (One Pass Dynamic Programming), so as to be a practical combined recognition system. Experimental results show that the recognition rates for voice phoneme, voice word, cursive character grapheme, and cursive character word are 51.65%, 88.6%, 85.3%, and 85.6%, respectively, when not using any language models. It demonstrates the efficiency of the proposed system.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.8
• /
• pp.1456-1462
• /
• 2008

In general, it is known a single biometric-based personal authentication has limitation to improve recognition rate due to weakness of individual recognition scheme. The recognition rate of face recognition system can be reduced by environmental factor such as illumination, while speaker verification system does not perform well with added surrounding noise. In this paper, a multi-modal biometric system composed of face and voice recognition system is proposed in order to improve the performance of the individual authentication system. The proposed empirical weight sum rule based on the reliability of the individual authentication system is applied to improve the performance of multi-modal biometrics. Since the proposed system is implemented using JAVA applet with security function, it can be utilized in the field of user authentication on the generic Web.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.7
• /
• pp.1-7
• /
• 2021

In this paper, we propose a feature selection technique for multi-label classification. Many existing feature selection techniques have selected features by calculating the relation between features and labels such as a mutual information scale. However, since the mutual information measure requires a joint probability, it is difficult to calculate the joint probability from an actual premise feature set. Therefore, it has the disadvantage that only a few features can be calculated and only local optimization is possible. Away from this regional optimization problem, we propose a feature selection technique that constructs a low-rank space in the entire given feature space and selects features with sparsity. To this end, we designed a regression-based objective function using Nuclear norm, and proposed an algorithm of gradient descent method to solve the optimization problem of this objective function. Based on the results of multi-label classification experiments on four data and three multi-label classification performance, the proposed methodology showed better performance than the existing feature selection technique. In addition, it was showed by experimental results that the performance change is insensitive even to the parameter value change of the proposed objective function.

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.9
• /
• pp.99-112
• /
• 2019

Cracks on pavement surfaces are critical signs and symptoms of the degradation of pavement structures. Image-based pavement crack detection is a challenging problem due to the intensity inhomogeneity, topology complexity, low contrast, and noisy texture background. In this paper, we address the problem of pavement crack detection and segmentation at pixel-level based on a Deep Neural Network (DNN) using gray-scale images. We propose a novel DNN architecture which contains a modified U-net network and a high-level features network. An important contribution of this work is the combination of these networks afforded through the fusion layer. To the best of our knowledge, this is the first paper introducing this combination for pavement crack segmentation and detection problem. The system performance of crack detection and segmentation is enhanced dramatically by using our novel architecture. We thoroughly implement and evaluate our proposed system on two open data sets: the Crack Forest Dataset (CFD) and the AigleRN dataset. Experimental results demonstrate that our system outperforms eight state-of-the-art methods on the same data sets.

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.6
• /
• pp.17-25
• /
• 2019

Accurate tracking of the sinkholes that are appearing frequently now is an important method of protecting human and property damage. Although many sinkhole detection systems have been proposed, it is still far from completely solved especially in-depth area. Furthermore, detection of sinkhole algorithms experienced the problem of unstable result that makes the system difficult to fire a warning in real-time. In this paper, we proposed a method of sinkhole tracking by deep learning and data association, that takes advantage of the recent development of CNN transfer learning. Our system consists of three main parts which are binary segmentation, sinkhole classification, and sinkhole tracking. The experiment results show that the sinkhole can be tracked in real-time on the dataset. These achievements have proven that the proposed system is able to apply to the practical application.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.1
• /
• pp.30-40
• /
• 2022

Electrocardiogram (ECG) signal's shape and characteristic varies through each individual, so it is difficult to classify with one neural network. It is difficult to classify the given data directly, but if corresponding normal beat is given, it is relatively easy and accurate to classify the beat by comparing two beats. In this study, we classify the ECG signal by generating the reference normal beat through the template cluster, and combining with the input ECG signal. It is possible to detect abnormal beats of various individual's records with one neural network by learning and classifying with the imaged ECG beats which are combined with corresponding reference normal beat. Especially, various neural networks, such as GoogLeNet, ResNet, and DarkNet, showed excellent performance when using the comparative learning. Also, we can confirmed that GoogLeNet has 99.72% sensitivity, which is the highest performance of the three neural networks.