• Journal of the Korea Society of Computer and Information
• /
• v.24 no.9
• /
• pp.21-27
• /
• 2019

In this paper, we propose a deep auto-encoder-based pipe leak detection (PLD) technique from time-series acoustic data collected by microphone sensor nodes. The key idea of the proposed technique is to learn representative features of the leak-free state using leak-free time-series acoustic data and the deep auto-encoder. The proposed technique can be used to create a PLD model that detects leaks in the pipeline in an unsupervised learning manner. This means that we only use leak-free data without labeling while training the deep auto-encoder. In addition, when compared to the previous supervised learning-based PLD method that uses image features, this technique does not require complex preprocessing of time-series acoustic data owing to the unsupervised feature extraction scheme. The experimental results show that the proposed PLD method using the deep auto-encoder can provide reliable PLD accuracy even considering unsupervised learning-based feature extraction.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.55 no.4
• /
• pp.202-206
• /
• 2006

In this paper, an auto detection method of stator winding fault of small induction motor is suggested. The Park's vector pattern which is obtained from 3-phase current signal by d-q transforming, is very good to detect winding fault. Comparing the Park's vector pattern of testing motor with its of healthy motor, the Park's vector pattern of fault motor is became an ellipse and the asymmetry is increased by the winding fault series. So for detecting the dis-symmetry, id-filtered function, Min-value, and Max-value are suggested for auto detecting. Using LabVIEW programing, 3-phase healthy motor and several kind of winding fault motors are tested and the test results are shown that the suggested method can gives us a possibility of an auto detecting winding fault.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.1
• /
• pp.25-32
• /
• 2018

In this paper, we propose an edge detection algorithm for auto focus of infrared camera. We designed and implemented the edge detection of infrared image by using a spatial filter on FPGA. The infrared camera should be designed to minimize the image processing time and usage of hardware resource because these days surveillance systems should have the fast response and be low size, weight and power. we applied the $3{\times}3$ mask filter which has an advantage of minimizing the usage of memory and the propagation delay to process filtering. When we applied Laplacian filter to extract contour data from an image, not only edge components but also noise components of the image were extracted by the filter. These noise components make it difficult to determine the focus state. Also a bad pixel of infrared detector causes a problem in detecting the edge components. So we propose an adaptive edge detection filter that is a method to extract only edge components except noise components of an image by analyzing a variance of pixel data in $3{\times}3$ memory area. And we can detect the bad pixel and replace it with neighboring normal pixel value when we store a pixel in $3{\times}3$ memory area for filtering calculation. The experimental result proves that the proposed method is effective to implement the edge detection for auto focus in infrared camera.

• IEIE Transactions on Smart Processing and Computing
• /
• v.5 no.2
• /
• pp.94-99
• /
• 2016

Application layer attacks have for years posed an ever-serious threat to network security, since they always come after a technically legitimate connection has been established. In recent years, cyber criminals have turned to fully exploiting the web as a medium of communication to launch a variety of forbidden or illicit activities by spreading malicious automated software (auto-ware) such as adware, spyware, or bots. When this malicious auto-ware infects a network, it will act like a robot, mimic normal behavior of web access, and bypass the network firewall or intrusion detection system. Besides that, in a private and large network, with huge Hypertext Transfer Protocol (HTTP) traffic generated each day, communication behavior identification and classification of auto-ware is a challenge. In this paper, based on a previous study, analysis of auto-ware communication behavior, and with the addition of new features, a method for classification of HTTP auto-ware communication is proposed. For that, a Not Only Structured Query Language (NoSQL) database is applied to handle large volumes of unstructured HTTP requests captured every day. The method is tested with real HTTP traffic data collected through a proxy server of a private network, providing good results in the classification and detection of suspicious auto-ware web access.

• Journal of Electrical Engineering and Technology
• /
• v.12 no.4
• /
• pp.1406-1416
• /
• 2017

In complex and large-scale industries, properly designed fault detection and identification (FDI) systems considerably improve safety, reliability and availability of target processes. In thermal power plants (TPPs), generating units operate under very dangerous conditions; system failures can cause severe loss of life and property. In this paper, we propose a bagged auto-associative kernel regression (AAKR)-based FDI approach for steam boilers in TPPs. AAKR estimates new query vectors by online local modeling, and is suitable for TPPs operating under various load levels. By combining the bagging method, more stable and reliable estimations can be achieved, since the effects of random fluctuations decrease because of ensemble averaging. To validate performance, the proposed method and comparison methods (i.e., a clustering-based method and principal component analysis) are applied to failure data due to water wall tube leakage gathered from a 250 MW coal-fired TPP. Experimental results show that the proposed method fulfills reasonable false alarm rates and, at the same time, achieves better fault detection performance than the comparison methods. After performing fault detection, contribution analysis is carried out to identify fault variables; this helps operators to confirm the types of faults and efficiently take preventive actions.

• The Journal of the Acoustical Society of Korea
• /
• v.36 no.4
• /
• pp.229-237
• /
• 2017

In this paper we propose an auto tonal detection method which exploits short term stationary when targets located in a detection beam area and then additional methods are proposed in order to reduce the computational complexity of the proposed method. The proposed method is adaptive to input signals and robust against interference caused by multiple targets because it compares an expected value of input signals with a threshold value which are estimated from a single beam while signals are keep stationary. The performances of the proposed methods are evaluated using by simulated data and acquired data from real ocean. The proposed method has shown better performance than conventional CFAR (Constant False Alarm Rate) methods.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.588-591
• /
• 2004

In this paper, we propose a ground plane detection algorithm, using a new image processing method (IPD). To extract the ground plane from the color image acquired by monocular camera, we use a new identical pixel detection method (IPD) and an edge detection method. This IPD method decides whether the pixel is identical with the ground plane pixel or not. The IPD method needs the reference area and its performance depends on the reference area size. So we propose the reference area auto-expanding algorithm in accordance with situation. And we evaluated the proposed algorithm by the experiments in the various environments. From the experiments results, we know that the proposed algorithm is efficient in the real indoor environment.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.2
• /
• pp.9-17
• /
• 2023

In this paper, we propose a one-class vibration anomaly detection system for bearing defect diagnosis. In order to reduce the economic and time loss caused by bearing failure, an accurate defect diagnosis system is essential, and deep learning-based defect diagnosis systems are widely studied to solve the problem. However, it is difficult to obtain abnormal data in the actual data collection environment for deep learning learning, which causes data bias. Therefore, a one-class classification method using only normal data is used. As a general method, the characteristics of vibration data are extracted by learning the compression and restoration process through AutoEncoder. Anomaly detection is performed by learning a one-class classifier with the extracted features. However, this method cannot efficiently extract the characteristics of the vibration data because it does not consider the frequency characteristics of the vibration data. To solve this problem, we propose an AutoEncoder model that considers the frequency characteristics of vibration data. As for classification performance, accuracy 0.910, precision 1.0, recall 0.820, and f1-score 0.901 were obtained. The network design considering the vibration characteristics confirmed better performance than existing methods.

• Proceedings of the KIEE Conference
• /
• 2006.10c
• /
• pp.229-231
• /
• 2006

NLPCA(Nonlinear Principal Component Analysis) is a novel technique for multivariate data analysis, similar to the well-known method of principal component analysis. NLPCA operates by a feedforward neural network called AANN(Auto Associative Neural Network) which performs the identity mapping. In this work, a sensor fault detection system based on NLPCA is presented. To verify its applicability, simulation study on the data supplied from sensor network is executed.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.12
• /
• pp.2547-2554
• /
• 2009

In this paper, we proposed a face detection algorithm and a hardware implementation method for ROI(Region Of Interest) of AF(Auto Focusing). We used face features in skin regions of YCbCr color space for face detection. The face features are the number of skin pixels in face regions, edge pixels in eye regions, and shadow pixels in lip regions. The each feature was statistically selected by 2,000 sample pictures of face. The proposed algorithm detects two faces that are closer center of the image for considering the effectiveness of hardware resource. The detected faces are displayed by rectangle for ROI of AF, and the rectangles are represented by positions in the image about starting point and ending point of the rectangles. The proposed face detection method was verified by using FPGA boards and mobile phone camera sensor.