• Proceedings of the KSRS Conference
• /
• 2005.10a
• /
• pp.233-235
• /
• 2005

This paper presents an unsupervised change detection methodology designed for the detection of landslide areas. The proposed methodology consists of two analytical steps: one for multi-temporal segmentation and the other for automatic selection of thresholding values. By considering the conditions of landslide occurrences and the spectral behavior of multi-temporal remote sensing images, some specific procedures are included in the analytical steps mentioned above. The effectiveness and applicability of the methodology proposed here were illustrated by a case study of the Gangneung area, Korea. The case study demonstrated that the proposed methodology could detect about $83\%$ of landslide occurrences.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.11
• /
• pp.3445-3461
• /
• 2000

Although many different intrusion detectionsystems have been developed there have not been enough researches on the methodology for evaluating these intrusion delection systems. With this understanding,in this paper we present a methodology for evaluating infrusion detection systems from the view point of performance and robustness, both of which are considered the most important criteria Current research on evaluating the performance f intrusion detection systems mostly foduson the in issuse detection but not on the anormaly detection. Regarding evalieting robustness it is not easy to apply off -line methodologies and methods for testing robustness hae not been proposed in on -line methodolomes, In this paper we provide an systematic way of classifyin and generating anomalies and using this reult, present an methodology for evaluating the pertormance of intrusion detection systems in detecting anomaalies ans well as misuses . Moreover, ww study the factors that can damage the robustness of intrusion detection systems and suggest an methodology for assessing the robustness of intrusion detection systems.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10b
• /
• pp.255-260
• /
• 1992

It is reported on the methodology of signal detection in noise which is based on a comparison of statistical parameters of observation sample from region of frequency-time noise space where a signal may be present and observation sample from region of this noise space and it is known a priori about the latter that the signal is absent in this region.

• Journal of Korean Institute of Industrial Engineers
• /
• v.9 no.1
• /
• pp.33-43
• /
• 1983

This paper develops a general methodology for evaluating inspection (mainly, safety inspection) policies, standard and regulations in a complex system with lots of components. Based on practically available data, this methodology enable planners and regulators to estimate the costs and effectiveness of different inspection policies when applied under different system working conditions. As main tools, the concepts of detection probability and earliness of detection are developed in this paper.

• 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.

• IEIE Transactions on Smart Processing and Computing
• /
• v.5 no.6
• /
• pp.390-394
• /
• 2016

This paper presents a new comparison method for haze-removal algorithms in next-generation automotive systems. Compared to previous peak signal-to-noise ratio-based comparisons, which measure similarity, the proposed modulation transfer function-based method checks sharpness to select a more suitable haze-removal algorithm for lane detection. Among the practical filtering schemes used for a haze-removal algorithm, experimental results show that Gaussian filtering effectively preserves the sharpness of road images, enhancing lane detection accuracy.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.101-125
• /
• 2022

Recently, with the development of computing technology and the improvement of the cloud environment, deep learning technology has developed, and attempts to apply deep learning to various fields are increasing. A typical example is anomaly detection, which is a technique for identifying values or patterns that deviate from normal data. Among the representative types of anomaly detection, it is very difficult to detect a contextual anomaly that requires understanding of the overall situation. In general, detection of anomalies in image data is performed using a pre-trained model trained on large data. However, since this pre-trained model was created by focusing on object classification of images, there is a limit to be applied to anomaly detection that needs to understand complex situations created by various objects. Therefore, in this study, we newly propose a two-step pre-trained model for detecting abnormal situation. Our methodology performs additional learning from image captioning to understand not only mere objects but also the complicated situation created by them. Specifically, the proposed methodology transfers knowledge of the pre-trained model that has learned object classification with ImageNet data to the image captioning model, and uses the caption that describes the situation represented by the image. Afterwards, the weight obtained by learning the situational characteristics through images and captions is extracted and fine-tuning is performed to generate an anomaly detection model. To evaluate the performance of the proposed methodology, an anomaly detection experiment was performed on 400 situational images and the experimental results showed that the proposed methodology was superior in terms of anomaly detection accuracy and F1-score compared to the existing traditional pre-trained model.

• Computational Structural Engineering
• /
• v.7 no.3
• /
• pp.153-166
• /
• 1994

A methodology to assess damage prediction accuracy as a function of model uncertainty in structures is presented. In the first part, a theory of approach is outlined. First, a damage detection algorithm to locate and size damage in structures using few modal responses of the structures is summarized. Next, methods to quantify model uncertainty and the damage detection accuracy are formulated. In the second part, a methodology to assess the effect of model uncertainty on the damage detection accuracy of real structures is designed. In the last part, the feasibility of the assessment methodology is demonstrated by using a plate-girder bridge for which only information on a single mode is available.

• Structural Engineering and Mechanics
• /
• v.42 no.3
• /
• pp.425-448
• /
• 2012

In this paper, a new damage detection and quantification method has been presented to perform detection and quantification of structural damage under ambient vibration loadings. To extract modal properties of the structural system under ambient excitation, natural excitation technique (NExT) and eigensystem realization algorithm (ERA) are employed. Sensitivity matrices of the dynamic residual force vector have been derived and used in the parameter subset selection method to identify multiple damaged locations. In the sequel, the steady state genetic algorithm (SSGA) is used to determine quantified levels of the identified damage by minimizing errors in the modal flexibility matrix. In this study, performance of the proposed damage detection and quantification methodology is evaluated using a finite element model of a truss structure with considerations of possible experimental errors and noises. A series of numerical examples with five different damage scenarios including a challengingly small damage level demonstrates that the proposed methodology can efficaciously detect and quantify damage under noisy ambient vibrations.

• Journal of Information Technology Applications and Management
• /
• v.15 no.1
• /
• pp.243-260
• /
• 2008

To prevent low yields in the semiconductor industry is crucial to the success of that industry. However, to prevent low yields is difficult because of too many factors to affect yield variation and their complex relation in the semiconductor manufacturing process. This study presents a new efficient detection methodology for detecting abnormal yields including high and low yields, which can forecast the yield level of a production unit (namely a lot) based on yield-related feature variables' behaviors. In the methodology, we use C5.0 to identify the yield-related feature variables that are the combination of correlated process variables associated with yield, use SOM (Self-Organizing Map) neural networks to extract and classify significant patterns of past abnormal yield lots and finally use C5.0 to generate classification rules for detecting abnormal yield lot. We illustrate the effectiveness of our methodology using a semiconductor manufacturing company's field data.