• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.251-266
• /
• 2022

Structural Health Monitoring (SHM) of critical infrastructure comprises a major pillar of maintenance management, shielding public safety and economic sustainability. Although SHM is usually associated with data-driven metrics and thresholds, expert judgement is essential, especially in cases where erroneous predictions can bear casualties or substantial economic loss. Considering that visual inspections are time consuming and potentially subjective, artificial-intelligence tools may be leveraged in order to minimize the inspection effort and provide objective outcomes. In this context, timely detection of sensor malfunctioning is crucial in preventing inaccurate assessment and false alarms. The present work introduces a sensor-fault detection and interpretation framework, based on the well-established support-vector machine scheme for anomaly detection, combined with a coalitional game-theory approach. The proposed framework is implemented in two datasets, provided along the 1st International Project Competition for Structural Health Monitoring (IPC-SHM 2020), comprising acceleration and cable-load measurements from two real cable-stayed bridges. The results demonstrate good predictive performance and highlight the potential for seamless adaption of the algorithm to intrinsically different data domains. For the first time, the term "decision trajectories", originating from the field of cognitive sciences, is introduced and applied in the context of SHM. This provides an intuitive and comprehensive illustration of the impact of individual features, along with an elaboration on feature dependencies that drive individual model predictions. Overall, the proposed framework provides an easy-to-train, application-agnostic and interpretable anomaly detector, which can be integrated into the preprocessing part of various SHM and condition-monitoring applications, offering a first screening of the sensor health prior to further analysis.

• Journal of the Korean Society for Aviation and Aeronautics
• /
• v.32 no.3
• /
• pp.69-78
• /
• 2024

This study evaluates the detection and utilization of wind shear using data from the Low-Level Wind Shear Alert System (LLWAS) and the Aerodrome Meteorological Observation System (AMOS) for the year 2023 at Incheon International Airport. A comparison of wind shear occurrence days revealed that LLWAS recorded 57 days, the reproduced LLWAS recorded 84 days, and AMOS recorded 163 days, with AMOS and the reproduced LLWAS showing higher occurrences. Performance metrics, including Probability of Detection (POD), False Alarm Ratio (FAR), Critical Success Index (CSI), and True Skill Statistic (TSS), were analyzed to evaluate detection capabilities. For the reproduced LLWAS, most wind shear events were detected, but the FAR was high, indicating lower performance. AMOS detected about 50% of actual wind shear events, with a lower FAR than the reproduced LLWAS but still relatively high. To improve detection performance, optimal thresholds for wind shear warnings were analyzed and adjusted, resulting in an increase in the CSI from 0.53 to 0.68 for the reproduced LLWAS and from 0.25 to 0.28 for AMOS. By adjusting the wind shear warning thresholds, the balance between POD and FAR was improved, confirming the potential for ground-based equipment to issue wind shear warnings effectively.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.4
• /
• pp.380-386
• /
• 2002

Conventional post-processing as like confidence measure (CM) proposed by Rahim calculates phones' CM using the likelihood between phoneme model and anti-model, and then word's CM is obtained by averaging phone-level CMs[1]. In conventional method, CMs of some specific keywords are tory low and they are usually rejected. The reason is that statistics of phone-level CMs are not consistent. In other words, phone-level CMs have different probability density functions (pdf) for each phone, especially sri-phone. To overcome this problem, in this paper, we propose normalized confidence measure. Our approach is to transform CM pdf of each tri-phone to the same pdf under the assumption that CM pdfs are Gaussian. For evaluating our method we use common keyword spotting system. In that system context-dependent HMM models are used for modeling keyword utterance and contort-independent HMM models are applied to non-keyword utterance. The experiment results show that the proposed NCM reduced FAR (false alarm rate) from 0.44 to 0.33 FA/KW/HR (false alarm/keyword/hour) when MDR is about 8%. It achieves 25% improvement of FAR.

• Journal of the Korea Society for Simulation
• /
• v.20 no.4
• /
• pp.31-40
• /
• 2011

Sensor nodes are easily exposed to malicious attackers by physical attacks. The attacker can generate various attacks using compromised nodes in a sensor network. The false report generating application layers injects the network by the compromised node. If a base station has the injected false report, a false alarm also occurs and unnecessary energy of the node is used. In order to defend the attack, a statistical en-route filtering method is proposed to filter the false report that goes to the base station as soon as possible. A path renewal method, which improves the method, is proposed to maintain a detection ability of the statistical en-route filtering method and to consume balanced energy of the node. In this paper, we proposed the secure path cycle method to consume effective energy for a path renewal. To select the secure path cycle, the base station determines through hop counts and the quantity of report transmission by an evaluation function. In addition, three methods, which are statistical en-route filter, path selection method, and path renewal method, are evaluated with our proposed method for efficient energy use. Therefore, the proposed method keeps the secure path and makes the efficiency of energy consumption high.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.10 no.6 s.52
• /
• pp.79-91
• /
• 2006

A new methodology of guided wave based nondestructive testing (NDT) is developed to detect crack damage in civil infrastructures such as steel bridges without using prior baseline data. In conventional guided wave based techniques, damage is often identified by comparing the "current" data obtained from a potentially damaged condition of a structure with the "past" baseline data collected at the pristine condition of the structure. However, it has been reported that this type of pattern comparison with the baseline data can lead to increased false alarms due to its susceptibility to varying operational and environmental conditions of the structure. To develop a more robust damage diagnosis technique, a new concept of NDT is conceived so that cracks can be detected without direct comparison with previously obtained baseline data. The proposed NDT technique utilizes the polarization characteristics of the piezoelectric wafers attached on the both sides of the thin metal structure. Crack formation creates Lamb wave mode conversion due to a sudden change in the thickness of the structure. Then, the proposed technique instantly detects the appearance of the crack by extracting this mode conversion from the measured Lamb waves even at the presence of changing operational and environmental conditions. Numerical and experimental results are presented to demonstrate the applicability of the proposed technique to crack detection.

• Journal of Food Hygiene and Safety
• /
• v.33 no.1
• /
• pp.65-70
• /
• 2018

Monitoring of Listeria monocytogenes, the causative agent of listeriosis, in food is inportant for public health. The Korean Food Standards Codex has adopted a 'zero-tolerance' policy for L. monocytogenes. The standard detection method of L. monocytogenes is based on enrichment. Thus, proper enrichment methods need to be instituted to ensure quality control of the detection procedures. In this study, the growth of L. monocytogenes and Listeria innocua as a mixed culture in Listeria enrichment broth (LEB) was monitored during artificial contamination of enrichment culture. We confirmed competitive growth or interspecies inhibitory activity of L. monocytogenes and L. innocua. Interspecies growth differences and the inhibitory activity of different inoculation and mixtures L. innocua against L. monocytogenes were examined. The concentration of L. monocytogenes must be 2.0 log CFU/mL or more than L. innocua to grow better than L. innocua. It is known that Listeria spp. and L. monocytogenes show growth difference during LEB, resulting in the risk of false-negative results. The inhibition of L. monocytogenes by L. innocua was always observed when present at lower concentrations. However, it was confirmed that L. innocua suppressed when L. monocytogenes was present at a higher concentration. Therefore if a mixture of Listeria spp. is present, detecting L. monocytogenes is difficult. Thus, a new enrichment broth to improve the detection rate of L. monocytogenes is needed.

• Asian-Australasian Journal of Animal Sciences
• /
• v.33 no.10
• /
• pp.1683-1690
• /
• 2020

Objective: The rapid and reliable detection of the African swine fever virus (ASFV) plays an important role in emergency control and preventive measures of ASF. Some methods have been recommended by FAO/OIE to detect ASFV in clinical samples, including realtime polymerase chain reaction (PCR). However, mismatches in primer and probe binding regions may cause a false-negative result. Here, a slight modification in probe sequence has been conducted to improve the qualification of real-time PCR based on World Organization for Animal Health (OIE) protocol for accurate detection of ASFV in field samples in Vietnam. Methods: Seven positive confirmed samples (four samples have no mismatch, and three samples contained one mutation in probe binding sites) were used to establish novel real-time PCR with slightly modified probe (Y = C or T) in comparison with original probe recommended by OIE. Results: Both real-time PCRs using the OIE-recommended probe and novel modified probe can detect ASFV in clinical samples without mismatch in probe binding site. A high correlation of cycle quantification (Cq) values was observed in which Cq values obtained from both probes arranged from 22 to 25, suggesting that modified probe sequence does not impede the qualification of real-time PCR to detect ASFV in clinical samples. However, the samples with one mutation in probe binding sites were ASFV negative with OIE recommended probe but positive with our modified probe (Cq value ranked between 33.12-35.78). Conclusion: We demonstrated for the first time that a mismatch in probe binding regions caused a false negative result by OIE recommended real-time PCR, and a slightly modified probe is required to enhance the sensitivity and obtain an ASF accurate diagnosis in field samples in Vietnam.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.10
• /
• pp.2461-2469
• /
• 1997

Target detection and recognition problems, in which neural networks are widely used, require translation invariant and real-time processing in addition to the requirements that general pattern recognition problems need. This paper presents a novel architecture that meets the requirements and explains effective methodology to train the network. The proposed neural network is an architectural extension of the shared-weight neural network that is composed of the feature extraction stage followed by the pattern recognition stage. Its feature extraction stage performs correlational operation on the input with a weight kernel, and the entire neural network can be considered a nonlinear correlation filter. Therefore, the output of the proposed neural network is correlational plane with peak values at the location of the target. The architecture of this neural network is suitable for implementing with parallel or distributed computers, and this fact allows the application to the problems which require realtime processing. Net training methodology to overcome the problem caused by unbalance of the number of targets and non-targets is also introduced. To verify the performance, the proposed network is applied to detection and recognition problem of a specific automobile driving around in a parking lot. The results show no false alarms and fast processing enough to track a target that moves as fast as about 190 km per hour.

• Korean Journal of Culture and Social Issue
• /
• v.14 no.4
• /
• pp.103-117
• /
• 2008

The polygraph test administered by the Korean Prosecutors Office for crime investigations customarily uses the score of -12 as the cutoff point separating the subjects who lie from those who tell the truth. The criterion used by the KPO is different from the one (-13) suggested by Backster (1963) who invented the particular method for lie detection. Based on the signal detection theory applied to the real polygraph test data obtained from real crime suspects by the KPO, the present study identified the score of -8 as an optimal criterion resulting in the highest overall accuracy of the polygraph test. The classification of the subjects with the score of -8 as the criterion resulted in the highest accuracy (83.17%) compared with the accuracies of classifications with the Backster's criterion (76.24%) and the KPO's criterion (80.20%). However, the new criterion was also found to result in more false-positive cases. Based on the results from the present study, it was recommended to use the score of -8 as the criterion when the overall accuracy is important but the score of -12 or -13 when avoiding false-positive is more important than securing the overall accuracy.

• The Korean Journal of Nuclear Medicine
• /
• v.35 no.5
• /
• pp.301-312
• /
• 2001

Purpose: Stomach cancer is one of the most common malignancies in Korea, but there is no report on FDG PET in patients with stomach cancer. We observed findings of FDG PET in patients with stomach cancer. Materials and Methods: In 13 patients with pre-operative stomach cancer, PET and CT were performed. Primary lesion and regional lymph nodes detection were aualyzed. Correlation between FDG uptake ratio and each prognostic factor of primary lesion was analyzed. In 19 patients diagnosed as recurrence or displaying suspicious symptoms, conventional work up including tumor marker and PET were performed. Recurrence detection of anastomotic site, distant metastasis, and tumor marker elevation were analyzed. Results: Sensitivity for primary lesion detection was 83.3% (CT 71.4%) and two submucosal lesions were undetected. FDG uptake ratio was variable and had no correlation with invasion-depth, size, Borrmann type, staging and differentiation. Sensitivity for regional lymph node detection was 58.3% (CT 58.3%) and the lesions less than 1cm were undetected. Sensitivity for recurrence detection was 100% but there were three false positives. Sensitivity for distant metastasis detection was 64.3% and significantly higher than that of conventional work-up (21.4%). Average of tumor marker level in patients who were confirmed as recurrence was higher than false positive. Conclusion: PET is more useful than conventional work up in distant metastasis detection when recurrence is suspected. In pre-operative stomach cancer, PET is comparable to CT for detection of primary lesion and regional lymph node metastasis and detection of distant metastasis requires further study.