• Proceedings of the Korea Society of Information Technology Applications Conference
• /
• 2005.11a
• /
• pp.153-156
• /
• 2005

Skin region detection in images is an important process in many computer vision applications targeting humans such as hand gesture recognition and face identification. It usually starts at a pixel-level, and involves a pre-process of color spae transformation followed by a classification process. A color space transformation is assumed to increase separability between skin classes and other classes, to increase similarity among different skin tones, and to bring a robust performance under varying imaging conditions, without any complicated analysis. In this paper, we examine if the color space transformation actually brings those benefits to the problem of skin region detection on a set of human hand images with different postures, backgrounds, people, and illuminations. Our experimental results indicate that color space transfomation affects the skin detection performance. Although the performance depends on camera and surround conditions, normalized [R, G, B] color space may be a good choice in general.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.28 no.1
• /
• pp.70-78
• /
• 2004

This paper presents study of efficiency of wavelet transformation for on-line chatter detection during hard fuming process. From comparison with other time series and statistical methods such as fast fourier transformation (FFT), Kurtosis and standard deviation (STD), wavelet transform is better than others in on-line chatter detection. With using wavelet function with pseudo frequency corresponding to chatter frequency, chatter could be detected more sensitively. And for both force signal from dynamometer and displacement signal from capacitance type cylindrical sensor (CCS), wavelet transform with DB2 function on level 4 could be well used for chatter detection in hard turning process.

• Structural Engineering and Mechanics
• /
• v.49 no.2
• /
• pp.183-202
• /
• 2014

A new method of multiple damage detection in beam like structures is introduced. The mode shapes of both healthy and damaged structures are used in damage detection process (DDP). Multiple Damage Localization Index Based on Mode Shapes (MDLIBMS) is presented as a criterion in detecting damaged elements. A finite element modeling of structures is used to calculate the mode shapes parameters. The main advantages of the proposed method are its simplicity, flexibility on the number of elements and so the accuracy of the damage(s) position(s), sensitivity to small damage extend, capability in prediction of required number of mode shapes and low sensitivity to noisy data. In fact, because of differential and comparative form of MDLIBMS, using noise polluted data doesn't have major effect on the results. This makes the proposed method a powerful one in damage detection according to measured mode shape data. Because of its flexibility, damage detection process in multi span bridge girders with non-prismatic sections can be done by this method. Numerical simulations used to demonstrate these advantages.

• Nuclear Engineering and Technology
• /
• v.52 no.12
• /
• pp.2687-2698
• /
• 2020

Most of the machine learning-based intrusion detection tools developed for Industrial Control Systems (ICS) are trained on network packet captures, and they rely on monitoring network layer traffic alone for intrusion detection. This approach produces weak intrusion detection systems, as ICS cyber-attacks have a real and significant impact on the process variables. A limited number of researchers consider integrating process measurements. However, in complex systems, process variable changes could result from different combinations of abnormal occurrences. This paper examines recent advances in intrusion detection algorithms, their limitations, challenges and the status of their application in critical infrastructures. We also introduce the discussion on the similarities and conflicts observed in the development of machine learning tools and techniques for fault diagnosis and cybersecurity in the protection of complex systems and the need to establish a clear difference between them. As a case study, we discuss special characteristics in nuclear power control systems and the factors that constraint the direct integration of security algorithms. Moreover, we discuss data reliability issues and present references and direct URL to recent open-source data repositories to aid researchers in developing data-driven ICS intrusion detection systems.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.11 no.2
• /
• pp.103-111
• /
• 2002

Out of all metal-cutting process, the hole-making process is the most widely used. It is estimated to be more than 30% of the total metal-cutting process. It is therefore desirable to monitor and detect drill wear during the hole-drilling process. One important aspect in controlling the drilling process is monitoring drill wear status. There are two systems, Basic system and Online system, to detect the drill wear. Basic system comprised of spindle rotational speed, feed rates, thrust torque and flank wear measured by tool microscope. Outline system comprised of spindle rotational speed feed rates, AE signal, flank wear area measured by computer vision, On-line monitoring system does not need to stop the process to inspect drill wear. Backpropagation neural networks (BPNs) were used for on-line detection of drill wear. The output was the drill wear state which was either usable or failure. This paper deals with an on-line drill wear monitoring system to fit the detection of the abnormal tool state.

• International Journal of Precision Engineering and Manufacturing
• /
• v.3 no.1
• /
• pp.5-12
• /
• 2002

The grinding process has been mainly used fur finishing metal products as final machining stage. But chatter vibration and bum of a workpiece have a bad effect on the machined surface and should be detected in modern grinding process. This paper deals with a fault detection of the cylindrical plunge grinding process by power parameters. During the grinding process the power signals of an induced motor were sampled and used to determine the relationship between fault and change of power parameters. A neural network was used far detecting the grinding fault and an influence of power parameters to the grinding fault was analyzed.

• Journal of Microbiology and Biotechnology
• /
• v.16 no.11
• /
• pp.1713-1719
• /
• 2006

A method for the simple, rapid, specific, and accurate detection of Bacillus anthracis spores was developed by employing specific capture peptides conjugated with fluorescent quantum dots (QDs). It was possible to distinguish B. anthracis spores from the spores of B. thuringiensis and B. cereus using these peptide-QD conjugates by flow cytometric and confocal laser scanning microscopic analyses. For more convenient high-throughput detection of B. anthracis spores, spectrofluorometric analysis of spore-peptide-QD conjugates was performed. B. anthracis spores could be detected in less than 1 h using this method. In order to avoid any minor yet false-positive signal caused by the presence of B. thuringiensis spores, the B-Negative peptide, which can only bind to B. thuringiensis, conjugated with another type of QD that fluoresces at different wavelength was also developed. In the presence of mixed B. anthracis and B. thuringiensis spores, the BABA peptide conjugated with QD525 and the B-Negative peptide conjugated with QD585 were able to bind to the former and the latter, specifically and respectively, thus allowing the clear detection of B. anthracis spores against B. thuringiensis spores by using two QD-labeling systems. This capture peptide-conjugated QD system should be useful for the detection of B. anthracis spores.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.45 no.4
• /
• pp.86-98
• /
• 2022

Recently, many studies have been conducted to improve quality by applying machine learning models to semiconductor manufacturing process data. However, in the semiconductor manufacturing process, the ratio of good products is much higher than that of defective products, so the problem of data imbalance is serious in terms of machine learning. In addition, since the number of features of data used in machine learning is very large, it is very important to perform machine learning by extracting only important features from among them to increase accuracy and utilization. This study proposes an anomaly detection methodology that can learn excellently despite data imbalance and high-dimensional characteristics of semiconductor process data. The anomaly detection methodology applies the LIME algorithm after applying the SMOTE method and the RFECV method. The proposed methodology analyzes the classification result of the anomaly classification model, detects the cause of the anomaly, and derives a semiconductor process requiring action. The proposed methodology confirmed applicability and feasibility through application of cases.

• Journal of Institute of Control, Robotics and Systems
• /
• v.6 no.12
• /
• pp.1113-1119
• /
• 2000

To operate a process plant safely and economically, process monitoring is very important. Process monitoring is the task to identify the state of the system from sensor data. Process monitoring includes data acquisition, regulatory control, data reconciliation, fault detection, etc. This research focuses on the data recon-ciliation using scale-space filtering and fault detection using functional-link associative neural networks. Scale-space filtering is a multi-resolution signal analysis method. Scale-space filtering can extract highest frequency factors(noise) effectively. But scale-space filtering has too large calculation costs and end effect problems. This research reduces the calculation cost of scale-space filtering by applying the minimum limit to the gaussian kernel. And the end-effect that occurs at the end of the signal of the scale-space filtering is overcome by using extrapolation related with the clustering change detection method. Nonlinear principal component analysis methods using neural network have been reviewed and the separately expanded functional-link associative neural network is proposed for chemical process monitoring. The separately expanded functional-link associative neural network has better learning capabilities, generalization abilities and short learning time than the exiting-neural networks. Separately expanded functional-link associative neural network can express a statistical model similar to real process by expanding the input data separately. Combining the proposed methods-modified scale-space filtering and fault detection method using the separately expanded functional-link associative neural network-a process monitoring system is proposed in this research. the usefulness of the proposed method is proven by its application a boiler water supply unit.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.6
• /
• pp.131-140
• /
• 2017

As video equipments such as CCTV are used for various purposes in fields of society, digital video data processing technology such as automatic motion detection is essential. In this paper, we proposed and implemented a more stable and accurate motion detection system based on background subtraction technique. We could improve the accuracy and stability of motion detection over existing methods by efficiently processing color information of digital image data. We divided the procedure of color information processing into each components of color information : brightness component, color component of color information and merge them. We can process each component's characteristics with maximum consideration. Our color information processing provides more efficient color information in motion detection than the existing methods. We improved the success rate of motion detection by our background update process that analyzed the characteristics of the moving background in the natural environment and reflected it to the background image.