• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.3
• /
• pp.86-97
• /
• 2021

As mechatronic systems have various, complex functions and require high performance, automatic fault detection is necessary for secure operation in manufacturing processes. For conducting automatic and real-time fault detection in modern mechatronic systems, multiple sensor signals are collected by internet of things technologies. Since traditional statistical control charts or machine learning approaches show significant results with unified and solid density models under normal operating states but they have limitations with scattered signal models under normal states, many pattern extraction and matching approaches have been paid attention. Signal discretization-based pattern extraction methods are one of popular signal analyses, which reduce the size of the given datasets as much as possible as well as highlight significant and inherent signal behaviors. Since general pattern extraction methods are usually conducted with a fixed size of time segmentation, they can easily cut off significant behaviors, and consequently the performance of the extracted fault patterns will be reduced. In this regard, adjustable time segmentation is proposed to extract much meaningful fault patterns in multiple sensor signals. By considering inflection points of signals, we determine the optimal cut-points of time segments in each sensor signal. In addition, to clarify the inflection points, we apply Savitzky-golay filter to the original datasets. To validate and verify the performance of the proposed segmentation, the dataset collected from an aircraft engine (provided by NASA prognostics center) is used to fault pattern extraction. As a result, the proposed adjustable time segmentation shows better performance in fault pattern extraction.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.10
• /
• pp.1-9
• /
• 2022

In this paper, we propose the Deep Learning-Based Companion Animal Abnormal Behavior Detection Service, which using video and sensor data. Due to the recent increase in households with companion animals, the pet tech industry with artificial intelligence is growing in the existing food and medical-oriented companion animal market. In this study, companion animal behavior was classified and abnormal behavior was detected based on a deep learning model using various data for health management of companion animals through artificial intelligence. Video data and sensor data of companion animals are collected using CCTV and the manufactured pet wearable device, and used as input data for the model. Image data was processed by combining the YOLO(You Only Look Once) model and DeepLabCut for extracting joint coordinates to detect companion animal objects for behavior classification. Also, in order to process sensor data, GAT(Graph Attention Network), which can identify the correlation and characteristics of each sensor, was used.

• Journal of The Korean Society of Clinical Toxicology
• /
• v.21 no.2
• /
• pp.128-134
• /
• 2023

Purpose: The Prescott nomogram has been utilized to forecast hepatotoxicity from acute acetaminophen poisoning. In developing countries, emergency medical centers lack the resources to report acetaminophen concentrations; thus, the commencement and cessation of treatment are based on the reported dose. This study investigated risk factors that can predict acetaminophen detection after 15 hours for safe treatment termination. Methods: Data were collected from an urban emergency medical center from 2010 to 2020. The study included patients ≥14 years of age with acute acetaminophen poisoning within 15 hours. The correlation between risk factors and detection of acetaminophen 15 hours after ingestion was evaluated using logistic regression, and the area under the curve (AUC) was calculated. Results: In total, 181 patients were included in the primary analysis; the median dose was 150.9 mg/kg and 35 patients (19.3%) had acetaminophen detected 15 hours after ingestion. The dose per weight and the time to visit were significant predictors for acetaminophen detection after 15 hours (odds ratio, 1.020 and 1.030, respectively). The AUCs were 0.628 for a 135 mg/kg cut-off value and 0.658 for a cut-off 450 minutes, and that of the combined model was 0.714 (sensitivity: 45.7%, specificity: 91.8%). Conclusion: Where acetaminophen concentrations are not reported during treatment following the UK guidelines, it is safe to start N-acetylcysteine immediately for patients who are ≥14 years old, visit within 15 hours after acute poisoning, and report having ingested ≥135 mg/kg. Additional N-acetylcysteine doses should be considered for patients visiting after 8 hours.

• Journal of Industrial Technology
• /
• v.38 no.1
• /
• pp.41-52
• /
• 2018

Recently, object detection is a critical function for any system that uses computer vision and is widely used in various fields such as video surveillance and self-driving cars. However, the conventional methods can not detect the objects clearly because of the dynamic background change in the beach. In this paper, we propose a new technique to detect humans correctly in the dynamic videos like shores. A new background modeling method that combines spatial GMM (Gaussian Mixture Model) and temporal GMM is proposed to make more correct background image. Also, the proposed method improve the accuracy of people detection by using SVM (Support Vector Machine) to classify people from the objects and KCF (Kernelized Correlation Filter) Tracker to track people continuously in the complicated environment. The experimental result shows that our method can work well for detection and tracking of objects in videos containing dynamic factors and situations.

• Journal of Biomedical Engineering Research
• /
• v.17 no.1
• /
• pp.109-120
• /
• 1996

Cardiac function is evaluated quantitatively using angiographic images via the analysis of the shape change or the heart wall boundaries. To kin with, boundary defection or ESLV(End Systolic Lert Ventricular) and EDLV(End Diastolic Left Ventricular) is essential for the quantitative analysis of cardiac function. The boundary detection methods proposed in the past were almost semi-automatic. Intervention by a knowledgeable human operator was still required Of con, manual tracing of the boundaries is currently used for subsequent analysis and diagnosis. This method would not cut excessive time, labor, and subjectivity associated with manual intervention by a human operator. EDLV images have noncontiguous and ambiguous edge signal on some boundary regions. In this paper, we propose a new method for automated detection of boundaries in noncontiguous and ambiguous EDLV images. The boundary detection scheme which based on a priori knowledge information is divided into two steps. The first step is to detect the candidate edge points of EDLV using ESLV boundaries. The second step is to correct detected boundaries of EDLV using the LV shape. We developed the algorithm of modifying EDLV boundaries defined adaptive modifier. We experimented the method proposed in this paper and compared our proposed method with the manual method in detecting boundaries of EDLV. In the areas within estimated boundaries of EDLV, the percentage of error was about 1.4%. We verified the useflilness and obtained the satisfying results througll the experiments of the proposed method.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.7 no.6
• /
• pp.64-71
• /
• 1998

This paper investigates the relationship between cutting conditions and Acoustic Emission(AE) signals; $AE_{avg}$, $AE_{rms}$, $AE_{mode}$$AE_{avg}$ and $AE_{rms}$ are increased as the increasing of cutting velocity and depth of cut respectively. The new parameters, derived from $AE_{avg}$ and $AE_{rms}$, which may be used for the in-process detection of tool wear is discussed. It is also known that $AE_{mode}$

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.32A no.12
• /
• pp.229-240
• /
• 1995

This paper proposes a new weighted random pattern testing technique to detect path delay faults in combinational logic circuits. When computing the probability of signal transition at primitive logic elements of CUT(Circuit Under Test) by the primary input, the proposed technique uses the information on the structure of CUT for initialization vectors and vectors generated by pseudo random pattern generator for test vectors. We can sensitize many paths by allocating a weight value on signal lines considering the difference of the levels of logic elements. We show that the proposed technique outperforms existing testing method in terms of test length and fault coverage using ISCAS '85 benchmark circuits. We also show that the proposed testing technique generates more robust test vectors for the longest and near-longest paths.

• Proceedings of the KIEE Conference
• /
• 2003.11b
• /
• pp.271-274
• /
• 2003

This paper presents a Built-in Current Sensor that detect defects in CMOS integrated circuits using the current testing technique. This scheme employs a cross-coupled connected PMOS transistors, it is used as a current comparator. Our proposed scheme is a negligible impart on the performance of the circuit undo. test (CUT). In addition, in the normal mode of the CUT not dissipation extra power, high speed detection time and applicable deep submicron process. The validity and effectiveness are verified through the HSPICE simulation on circuits with defects. The entire area of the test chip is $116{\times}65{\mu}m^2$. The BICS occupies only $41{\times}17{\mu}m^2$ of area in the test chip. The area overhead of a BICS versus the entire chip is about 9.2%. The chip was fabricated with Hynix $0.35{\mu}m$ 2-poly 4-metal N-well CMOS technology.

• International journal of advanced smart convergence
• /
• v.11 no.1
• /
• pp.1-10
• /
• 2022

Single-view reconstruction (SVR) is a fundamental method in computer vision. Often used for reconstructing human-made environments, the Manhattan world assumption presumes that planes in the real world exist in mutually orthogonal directions. Accordingly, this paper addresses an automatic SVR algorithm for Manhattan worlds. A method for estimating the directions of planes using graph-cut optimization is proposed. After segmenting an image from extracted line segments, the data cost function and smoothness cost function for graph-cut optimization are defined by considering the directions of the line segments and neighborhood segments. Furthermore, segments with the same depths are grouped during a depth-estimation step using a minimum spanning tree algorithm with the proposed weights. Experimental results demonstrate that, unlike previous methods, the proposed method can identify complex Manhattan structures of indoor and outdoor scenes and provide the exact boundaries and intersections of planes.

• Journal of KIISE:Software and Applications
• /
• v.29 no.11
• /
• pp.837-850
• /
• 2002

In this paper, we propose an EEG-based response time prediction method during a yes/no cognitive decision task. In the experimental task, a subject goes through responding of visual stimulus, understanding the given problem, controlling hand motions, and hitting a key. Considering the subject's varying brain activities, we model subjects' mental states with defining CT (cut time), ST (selection time), and RP (repeated period). Based on the assumption between ST and RT in the mental model, we predict subjects' response time by detection of selection time. To recognize the subjects' selection time ST, we extract 3 types of feature from the filtered brain waves at frequency bands of $\alpha$, $\beta$, ${\gamma}$ waves in 4 electrode pairs combined by spatial relationships. From the extracted features, we construct specific rules for each subject and meta rules including common factors in all subjects. Applying the ST detection rules to 8 subjects gives 83% success rates and also shows that the subjects will hit a key in 0.73 seconds after ST detected. To validate the detection rules and parameters, we test the rules for 2 subjects among 8 and discuss about the experimental results. We expect that the proposed detection method can be a basic technology for brain-computer-interface by combining with left/right hand movement or yes/no discrimination methods.