• Journal of the Korean Association of Geographic Information Studies
• /
• v.4 no.4
• /
• pp.1-7
• /
• 2001

Remotely sensed data is the most fundamental data in acquiring the GIS informations, and may be analyzed to extract useful thematic information. Multi-spectral classification is one of the most often used methods of information extraction. The actual multi-spectral classification may be performed using either supervised or unsupervised approaches. This paper analyze the effect of assigning clever initial values to image classes on the performance of parallelepiped classification algorithm, which is one of the supervised classification algorithms. First, we investigate the effect on serial computing model, then expand it on MIMD(Multiple Instruction Multiple Data) parallel computing model. On serial computing model, the performance of the parallel pipe algorithm improved 2.4 times at most and, on MIMD parallel computing model the performance improved about 2.5 times as clever initial values are assigned to image class. Through computer simulation we find that initial values of image class greatly affect the performance of parallelepiped classification algorithms, and it can be improved greatly when classes on both serial computing model and MIMD parallel computation model.

• Journal of Korea Multimedia Society
• /
• v.20 no.12
• /
• pp.1960-1969
• /
• 2017

There have been ongoing researches to identify and analyze the patterns of electric power IoT data inside sensor nodes to supplement the stable supply of power and the efficiency of energy consumption. This study set out to propose an analysis process for electric power IoT data with the K-means algorithm, which is an unsupervised learning technique rather than a supervised one. There are a couple of problems with the old K-means algorithm, and one of them is the selection of cluster number K in a heuristic or random method. That approach is proper for the age of standardized data. The investigator proposed an analysis process of selecting an automated cluster number K through principal component analysis and the space division of normal distribution and incorporated it into electric power IoT data. The performance evaluation results show that it recorded a higher level of performance than the old algorithm in the cluster classification and analysis of pitches and rolls included in the communication bodies of utility poles.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.8
• /
• pp.2555-2564
• /
• 2000

In this paper, a new unsupervised image segmentation algorithm is proposed. To model the contextual information presented in images, the characteristics of the Markov random fields (MRF) are utilized. Textured images are modeled as realizations of the stationary Gaussian MRF on a two-dimensional square lattice using the conditional autoregressive (CAR) equations with a second-order noncausal neighborhood. To detect boundaries, hypothesis tests over two masked areas are performed. Under the hypothesis, masked areas are assumed to belong to the same class of textures and CAR equation parameters are estimated in a minimum-mean-square-error (MMSE) sense. If the hypothesis is rejected, a measure of dissimilarity between two areas is accumulated on the rejected area. This approach produces potential edge maps. Using these maps, boundary detection can be performed, which resulting no micro edges. The performance of the proposed algorithm is evaluated by some experiments using real images as weB as synthetic ones. The experiments demonstrate that the proposed algorithm can produce satisfactorY segmentation without any a priori information.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.8 no.4
• /
• pp.185-194
• /
• 2013

Abstract- In distant-talking environments, speech recognition performance degrades significantly due to noise and reverberation. Recent work of Michael L. Selzer shows that in microphone array speech recognition, the word error rate can be significantly reduced by adapting the beamformer weights to generate a sequence of features which maximizes the likelihood of the correct hypothesis. In this approach, called Likelihood Maximizing Beamforming algorithm (Limabeam), one of the method to implement this Limabeam is an UnSupervised Limabeam(USL) that can improve recognition performance in any situation of environment. From our investigation for this USL, we could see that because the performance of optimization depends strongly on the transcription output of the first recognition step, the output become unstable and this may lead lower performance. In order to improve recognition performance of USL, some post-filter techniques can be employed to obtain more correct transcription output of the first step. In this work, as a post-filtering technique for first recognition step of USL, we propose to add a Wiener-Filter combined with Feature Weighted Malahanobis Distance to improve recognition performance. We also suggest an alternative way to implement Limabeam algorithm for Hidden Markov Network (HM-Net) speech recognizer for efficient implementation. Speech recognition experiments performed in real distant-talking environment confirm the efficacy of Limabeam algorithm in HM-Net speech recognition system and also confirm the improved performance by the proposed method.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.2C
• /
• pp.162-171
• /
• 2003

It is difficult to reuse the captured motion data, because the data has a difficulty in editing it. In this paper, a uniform posture mar (UPM) algorithm, one of unsupervised learning neural network is proposed to edit the captured motion data. Because it needs much less computational cost than other motion editing algorithms, it is adequate to apply in teal-time applications. The UPM algorithm prevents from generating an unreal posture in learning phase. It not only makes more realistic motion curves, but also contributes to making more natural motions. Above of all, it complements the weakness of the existing algorithm where the calculation quantity increases in proportion to increase the number of restricted condition to solve the problems of high order articulated body. In this paper, it is shown two applications as a visible the application instance of UPM algorithm. One is a motion transition editing system, the other is a inductive inverse kinematics system. This method could be applied to produce 3D character animation based on key frame method, 3D game, and virtual reality, etc.

• Journal of Korea Water Resources Association
• /
• v.33 no.4
• /
• pp.505-515
• /
• 2000

THe purpose of this study is to forecast of runoff hydrographs according to rainfall event in a stream. The neural network theory as a hydrologic blackbox model is used to solve hydrological problems. The Back-Propagation(BP) algorithm by the Levenberg-Marquardt(LM) techniques and Radial Basis Function(RBF) network in Neural Network(NN) models are used. Runoff hydrograph is forecasted in Bocheongstream basin which is a IHP the representative basin. The possibility of a simulation for runoff hydrographs about unlearned stations is considered. The results show that NN models are performed to effective learning for rainfall-runoff process of hydrologic system which involves a complexity and nonliner relationships. The RBF networks consist of 2 learning steps. The first step is an unsupervised learning in hidden layer and the next step is a supervised learning in output layer. Therefore, the RBF networks could provide rather time saved in the learning step than the BP algorithm. The peak discharge both BP algorithm and RBF network model in the estimation of an unlearned are a is trended to observed values.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.24 no.2
• /
• pp.141-148
• /
• 2011

This study presents the optimal grouping technique of columns which groups together columns of similar shortening trends to improve the efficiency of column shortening compensation. Here, Kohonen's self-organizing feature map which can classify patterns of input data by itself with unsupervised learning was used as the optimal grouping algorithm. The Kohonen network applied in this study is composed of two input neurons and variable output neurons, here the number of output neuron is equal to the column groups to be classified. In input neurons the normalized mean and standard deviation of shortening of each columns are inputted and in the output neurons the classified column groups are presented. The applicability of the proposed algorithm was evaluated by applying it to the two buildings where column shortening analyses had already been performed. The proposed algorithm was able to classify columns with similar shortening trends as one group, and from this we were able to ascertain the field-applicability of the proposed algorithm as the optimal grouping of column shortening.

• The Journal of Society for e-Business Studies
• /
• v.27 no.1
• /
• pp.1-13
• /
• 2022

With the development of network technologies, the security to protect organizational resources from internal and external intrusions and threats becomes more important. Therefore in recent years, the anomaly detection algorithm that detects and prevents security threats with respect to various security log events has been actively studied. Security anomaly detection algorithms that have been developed based on rule-based or statistical learning in the past are gradually evolving into modeling based on machine learning and deep learning. In this study, we propose a deep-autoencoder model that transforms LSTM-autoencoder as an optimal algorithm to detect insider threats in advance using various machine learning analysis methodologies. This study has academic significance in that it improved the possibility of adaptive security through the development of an anomaly detection algorithm based on unsupervised learning, and reduced the false positive rate compared to the existing algorithm through supervised true positive labeling.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.6
• /
• pp.445-452
• /
• 2019

In this study, parallel processing codes of k-means clustering algorithm were developed and implemented in a PC-cluster for unsupervised classification of large satellite images. We implemented intra-node code using multicores of CPU (Central Processing Unit) based on OpenMP (Open Multi-Processing), inter-nodes code using a PC-cluster based on message passing interface, and hybrid code using both. The PC-cluster consists of one master node and eight slave nodes, and each node is equipped with eight multicores. Two operating systems, Microsoft Windows and Canonical Ubuntu, were installed in the PC-cluster in turn and tested to compare parallel processing performance. Two multispectral satellite images were tested, which are a medium-capacity LANDSAT 8 OLI (Operational Land Imager) image and a high-capacity Sentinel 2A image. To evaluate the performance of parallel processing, speedup and efficiency were measured. Overall, the speedup was over N / 2 and the efficiency was over 0.5. From the comparison of the two operating systems, the Ubuntu system showed two to three times faster performance. To confirm that the results of the sequential and parallel processing coincide with the other, the center value of each band and the number of classified pixels were compared, and result images were examined by pixel to pixel comparison. It was found that care should be taken to avoid false sharing of OpenMP in intra-node implementation. To process large satellite images in a PC-cluster, code and hardware should be designed to reduce performance degradation caused by file I / O. Also, it was found that performance can differ depending on the operating system installed in a PC-cluster.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.17 no.2
• /
• pp.15-23
• /
• 2017

The person Re-identification is the most challenging part of computer vision due to the significant changes in human pose and background clutter with occlusions. The picture from non-overlapping cameras enhance the difficulty to distinguish some person from the other. To reach a better performance match, most methods use feature selection and distance metrics separately to get discriminative representations and proper distance to describe the similarity between person and kind of ignoring some significant features. This situation has encouraged us to consider a novel method to deal with this problem. In this paper, we proposed an enhanced recurrent neural network with three-tier hierarchical network for person re-identification. Specifically, the proposed recurrent neural network (RNN) model contain an iterative expectation maximum (EM) algorithm and three-tier Hierarchical network to jointly learn both the discriminative features and metrics distance. The iterative EM algorithm can fully use of the feature extraction ability of convolutional neural network (CNN) which is in series before the RNN. By unsupervised learning, the EM framework can change the labels of the patches and train larger datasets. Through the three-tier hierarchical network, the convolutional neural network, recurrent network and pooling layer can jointly be a feature extractor to better train the network. The experimental result shows that comparing with other researchers' approaches in this field, this method also can get a competitive accuracy. The influence of different component of this method will be analyzed and evaluated in the future research.