• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.1
• /
• pp.161-169
• /
• 1996

This paper presents a new search region prediction method using vector quantization for the motion estimation. We find motion vectors using the full search BMA from two successive frame images first. Then the motion vectors are used for training a codebook. The trained codebook is the predicted search region. We used the unsupervised neural network for VQ encoding and codebook design. A major advantage of formulating VQ as neural networks is that the large number of adaptive training algorithm that are used for neural networks can be applied to VQ. The proposed method reduces the computation and reduce the bits required to represent the motion vectors because of the smaller search points. The computer simulation results show the increased PSNR as compared with the other block matching algorithms.

• Proceedings of the KSRS Conference
• /
• 2008.03a
• /
• pp.195-200
• /
• 2008

The accuracy of classifying pixels in HIRIS images is usually degraded by noisy bands since noisy bands may deform the typical shape of spectral reflectance. Proposed in this paper is a statistical method for noisy band removal which mainly makes use of the correlation coefficients between bands. Considering each band as a random variable, the correlation coefficient measures the strength and direction of a linear relationship between two random variables. While the correlation between two signal bands is high, existence of a noisy band will produce a low correlation due to ill-correlativeness and undirectedness. The application of the correlation coefficient as a measure for detecting noisy bands is under a two-pass screening scheme. This method is independent of the prior knowledge of the sensor or the cause resulted in the noise. The classification in this experiment uses the unsupervised k-nearest neighbor algorithm in accordance with the well-accepted Euclidean distance measure and the spectral angle mapper measure. This paper also proposes a hierarchical combination of these measures for spectral matching. Finally, a separability assessment based on the between-class and within-class scatter matrices is followed to evaluate the performance.

• Journal of Internet Computing and Services
• /
• v.10 no.4
• /
• pp.45-54
• /
• 2009

In this paper we propose a new approach that recognizes the similar categories by weighting distinctive features. The approach is based on the PLSA that is one of the effective methods for the object categorization. PLSA is introduced from the information retrieval of text domain. PLSA, unsupervised method, shows impressive performance of category recognition. However, it shows relatively low performance for the similar categories which have the analog distribution of the features. In this paper, we consider the effective object categorization for the similar categories by weighting the mainly distinctive features. We present that the proposed algorithm, weighted PLSA, recognizes similar categories. Our method shows better results than the standard PLSA.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.12
• /
• pp.5643-5656
• /
• 2016

The detection accuracy of steganalysis depends on many factors, including the embedding algorithm, the payload size, the steganalysis feature space and the properties of the cover source. In practice, the cover source mismatch (CSM) problem has been recognized as the single most important factor negatively affecting the performance. To address this problem, we propose a new framework for blind, universal steganalysis which uses traditional steganalyst features. Firstly, cover images with the same statistical properties are searched from a reference image database as aided samples. The test image and its aided samples form a whole test set. Then, by assuming that most of the aided samples are innocent, we conduct outlier detection on the test set to judge the test image as cover or stego. In this way, the framework has removed the need for training. Hence, it does not suffer from cover source mismatch. Because it performs anomaly detection rather than classification, this method is totally unsupervised. The results in our study show that this framework works superior than one-class support vector machine and the outlier detector without considering the image retrieval process.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.43 no.1
• /
• pp.123-131
• /
• 2020

There have been a lot of studies in the past for the method of predicting the failure of a machine, and recently, a lot of researches and applications have been generated to diagnose the physical condition of the machine and the parts and to calculate the remaining life through various methods. Survival models are also used to predict plant failures based on past anomaly cycles. In particular, special machine that reflect the fluid flow and process characteristics of chemical plants are connected to hundreds or thousands of sensors, so there are not many factors that need to be considered, such as process and material data as well as application of derivative variables. In this paper, the data were preprocessed through time series anomaly detection based on unsupervised learning to predict the abnormalities of these special machine. Next, clustering results reflecting clustering-based data characteristics were applied to produce additional variables, and a learning data set was created based on the history of past facility abnormalities. Finally, the prediction methodology based on the supervised learning algorithm was applied, and the model update was confirmed to improve the accuracy of the prediction of facility failure. Through this, it is expected to improve the efficiency of facility operation by flexibly replacing the maintenance time and parts supply and demand by predicting abnormalities of machine and extracting key factors.

• Proceedings of the Korea Contents Association Conference
• /
• 2016.05a
• /
• pp.53-54
• /
• 2016

빅데이터란 대용량 데이터 활용 및 분석을 통해 가치 있는 정보를 추출하고, 이를 바탕으로 대응 방안 도출 또는 변화를 예측하는 기술을 의미한다. 그리고 빅데이터 분석에 활용되는 데이터인 페이스북과 같은 소셜 데이터, 유전자 발현과 같은 바이오 데이터, 항공망과 같은 지리정보 데이터들은 대용량 네트워크로 구성되어 있다. 네트워크 클러스터링은 서로 유사한 특성을 갖는 네트워크 내의 데이터들을 동일한 클러스터로 묶는 기법으로 네트워크 데이터를 분석하고 그 특성을 파악하는데 폭넓게 사용된다. 최근 빅데이터가 다양한 분야에서 활용되면서 방대한 양의 네트워크 데이터가 생성되고 있고, 이에 따라서 대용량 네트워크 데이터를 효율적으로 처리하는 클러스터링 기법의 중요성이 증가하고 있다. MCL(Markov Clustering) 알고리즘은 플로우 기반 무감독(unsupervised) 클러스터링 알고리즘으로 확장성이 우수해 다양한 분야에서 활용되고 있다. 하지만, MCL은 대용량 네트워크에 대해서는 많은 클러스터링 연산을 요구하며 너무 많은 클러스터를 생성하는 문제를 갖는다. 본 논문에서는 네트워크 압축을 기반으로 한 클러스터링 알고리즘을 제안함으로써 MCL보다 클러스터링 속도와 정확도를 향상시켰다. 또한, 희소행렬을 효율적으로 저장하는 CSC(Compressed Sparse Column) 자료구조와 MapReduce 기법을 제안한 클러스터링 알고리즘에 적용함으로써 대용량 네트워크에 대한 클러스터링 속도를 향상시켰다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10b
• /
• pp.1433-1436
• /
• 2000

클러스터링에 있어서 k-means[7], DBSCAN[2], CURE[4], ROCK[5], PAM[8], 같은 기존의 알고리즘은 원형이나 타원형 등의 어느 고정된 모양에 의해 클러스터를 결정한다. 만약 클러스터 하려는 데이터의 분포가 우연히 알고리즘의 결정된 모양과 일치하면 정확한 해를 얻을 수 있다. 하지만 자연적인 데이터의 분포에서는 발생하기 어렵다. 데이터의 형태를 추적하여 이러한 문제점을 해결한 CHAMELEON[1] 알고리즘이 최근에 발표되었다. 하지만 모양에는 독립적이나 데이터의 양이 증가함에 따라 소요되는 시간이 폭발적으로 증가한다. 이것은 기존의 마이닝 데이터들이 대용량이라는 것을 고려하면 현실에 적용하기 힘든 문제점이 있다. 이러한 문제점을 해결하기 위해 본 논문에서는 K-means[7]]를 이용한 대표를 선출하는 방법으로 CHAMELEON[1]의 문제점 개선(EF-CHAMELEON)을 시도하였으며 여러 자연적인 형태의 도형들은 아주 작은 원형들의 집합으로 구성 될 수 있다는 생각을 기본으로 잡음에 영향을 받지 않을 정도로 아주 작은 초기 다수의 소형 클러스터를 K-mean을 이용하여 구성하고 이를 다시 크러스터간의 상대적인 거리를 이용하여 다시 머지 하는 방법으로 모양에 의존적인 문제를 해결하며 비교사 학습(unsupervised learning)에 충실하기 위해 임계값을 적용 적정 단계에서 알고리즘을 멈추게 한 ADF 알고리즘을 소개한다. 실험 데이터는 기존의 여러 클러스터링 알고리즘이 판별 할 수 없었던 다양한 모양을 가지고있는 2차원 배열을 사용하여 ADF. CHAMELEON[1], EF-CHAMELEON,의 성능을 비교하였다.

• Journal of Electrical Engineering and Technology
• /
• v.9 no.4
• /
• pp.1426-1436
• /
• 2014

In this article, we present a new grey scale image segmentation method based on Fuzzy logic and bitplane techniques which combines the bits of different bitplanes of a pixel inorder to increase the segmentation quality and to get a more reliable and accurate segmentation result. The proposed segmentation approach is conceptually different and explores a new strategy. Infact, our technique consists in combining many realizations of the image together inorder to increase the information quality and to get an optimal segmented image. For segmentation, we proceed in two steps. In the first step, we begin by identifying the bitplanes that represent the lungs clearly. For this purpose, the intensity value of a pixel is separated into bitplanes. In the second step, segmentation values are assigned for each bitplane based on membership table. The segmented values of foreground are combined and the segmentation values of background are combined. The algorithm is demonstrated through the medical computed tomography (CT) images. The segmentation accuracy of the proposed method is compared with two existing techniques. Satisfactory segmentation results have been obtained showing the effectiveness and superiority of the proposed method.

• Journal of Korea Multimedia Society
• /
• v.19 no.9
• /
• pp.1647-1658
• /
• 2016

RGB+D database has been widely used in object recognition, object tracking, robot control, to name a few. While rapid advance of active depth sensing technologies allows for the widespread of indoor RGB+D databases, there are only few outdoor RGB+D databases largely due to an inherent limitation of active depth cameras. In this paper, we propose a novel method used to build outdoor RGB+D databases. Instead of using active depth cameras such as Kinect or LIDAR, we acquire a pair of stereo image using high-resolution stereo camera and then obtain a depth map by applying stereo matching algorithm. To deal with estimation errors that inevitably exist in the depth map obtained from stereo matching methods, we develop an approach that estimates confidence of depth maps based on unsupervised learning. Unlike existing confidence estimation approaches, we explicitly consider a spatial correlation that may exist in the confidence map. Specifically, we focus on refining confidence feature with the assumption that the confidence feature and resultant confidence map are smoothly-varying in spatial domain and are highly correlated to each other. Experimental result shows that the proposed method outperforms existing confidence measure based approaches in various benchmark dataset.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.2
• /
• pp.321-330
• /
• 2019

Prevention of corporate confidentiality leakage by insiders in enterprises is an essential task for the survival of enterprises. In order to prevent information leakage by insiders, companies have adopted security solutions, but there is a limit to effectively detect abnormal behavior of insiders with access privileges. In this study, we use the Unsupervised Learning algorithm of the machine learning technique to effectively and efficiently cluster the normal and abnormal access logs of the worker's work screen in the manufacturing information system, which includes the company's product manufacturing history and quality information. We propose an optimal feature selection model for anomaly detection by studying clustering methods.