• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.89-92
• /
• 2021

In a manufacturing fields, an abnormality or breakdown of equipment is a factor that causes product defects. Recently, with the spread of smart factory services, a lot of research to predict and prevent machine's failures is actively ongoing. However, there is a big difficulty in developing a classification model because the number of abnormal or failure data of the machine is severely smaller than normal data. In this paper, we present an algorithm for detecting abnormalities in an MCT at manufacturing work site depending on the differences between inputs and outputs of Autoencoder model and analyze its performance. The algorithm detects abnormalities using only features of normal data from manufacturing data of the MCT in which abnormal data does not exist.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2024.01a
• /
• pp.41-43
• /
• 2024

본 연구에서는 Vision Transformer 기반의 Anomaly Detection and Localization (VT-ADL) 모델에 초점을 맞추고, 손실 함수의 변경이 MVTec 데이터셋에 대한 이상 검출 및 지역화 성능에 미치는 영향을 비교 분석한다. 기존의 손실 함수를 KL Divergence와 Log-Likelihood Loss의 조합인 VAE Loss로 대체하여, 성능 변화를 심층적으로 조사했다. 실험을 통해 VAE Loss로의 전환은 VT-ADL 모델의 이상 검출 능력을 현저히 향상시키며, 특히 PRO-score에서 기존 대비 약 5%의 개선을 보였다는 점을 확인하였다. 이러한 결과는 손실 함수의 최적화가 VT-ADL 모델의 전반적인 성능에 중요한 영향을 미칠 수 있음을 시사한다. 또한, 이 연구는 Vision Transformer 기반 모델의 이상 검출과 지역화 작업에 있어서 손실 함수 선택의 중요성을 강조하며, 향후 관련 연구에 유용한 기준을 제공할 수 있을 것으로 기대된다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2024.01a
• /
• pp.47-48
• /
• 2024

Heating, Ventilating, and Air Conditioning(HVAC) 시스템은 난방(Heating), 환기(Ventilating), 공기조화(Air Conditioning)를 제공하는 공조시스템으로, 실내 환경의 온도, 습도 조절 및 지속적인 순환 및 여과를 통해 실내 공기 질을 개선한다. 이러한 HVAC 시스템에 이상이 생기는 경우 공기 여과율이 낮아지며, COVID-19와 같은 법정 감염병 예방에 취약해진다. 또한 장비의 과부하를 유발하여, 시스템의 효율성 저하 및 에너지 낭비를 불러올 수 있다. 따라서 본 논문에서는 HVAC 시스템의 이상 탐지 및 조기 조치를 위한 Transformer 기반 이상 탐지 기법의 적용을 제안한다. Transformer는 기존 시계열 데이터 처리를 위한 기법인 Recurrent Neural Network(RNN)기반 모델의 구조적 한계점을 극복함에 따라 Long Term Dependency 문제를 해결하고, 병렬처리를 통해 효율적인 Feature 추출이 가능하다. Transformer 모델이 HVAC 시스템의 이상 탐지에서 RNN 기반의 비교군 모델보다 약 1.31%의 향상을 보이며, Transformer 모델을 통한 HVAC의 이상 탐지에 효율적임을 확인하였다.

• Annual Conference of KIPS
• /
• 2024.10a
• /
• pp.277-280
• /
• 2024

Collaborative inference combines diverse features contributed by various agents to improve prediction accuracy. However, it is vulnerable to adversarial attacks, where attackers manipulate the model's predictions through non-consensual inputs. Since each participant operates within their unique feature space, defending against these attacks becomes particularly challenging. A recent study demonstrated that using an auto-encoder based on the underlying manifold can reduce the impact of malicious participants. However, our experiments observed that the recently proposed attack, in which malicious influences close to the manifold, may still pose a threat. To address this issue, we introduce a novel approach that leverages implicit redundancy across participants' feature spaces during the inference stage via participant-wise anomaly detection. We evaluate this approach on CIFAR10, CINIC10, Imagenette, Give-Me-Some-Credit, and Bank Marketing datasets. Extensive experiments and ablation studies show that RIP effectively mitigates adversarial attacks in the collaborative inference stage.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.6 no.2
• /
• pp.199-207
• /
• 2010

Recently, large amount of information in IDS(Intrusion Detection System) can be un manageable and also be mixed with false prediction error. In this paper, we propose a data mining methodology for IDS, which contains uncertainty based on training process and post-processing analysis additionally. Our system is trained to classify the existing attack for misuse detection, to detect the new attack pattern for anomaly detection, and to define border patter between attack and normal pattern. In experimental results show that our approach improve the performance against existing attacks and new attacks, from 0.62 to 0.84 about 35%.

• Journal of Electrical Engineering and Technology
• /
• v.7 no.3
• /
• pp.429-435
• /
• 2012

In this study, a real-time architecture for the detection of clock jumps in the GPS clock behavior is proposed. GPS satellite atomic clocks have characteristics of a second order polynomial in the long term showing sudden jumps occasionally. As satellite clock anomalies influence on GPS measurements which could deliver wrong position information to users as a result, it is required to develop a real time technique for the detection of the clock anomalies especially on the real-time GPS applications such as aviation. The proposed strategy is based on Teager Energy operator, which can be immediately detect any changes in the satellite clock bias estimated from GPS carrier phase measurements. The verification results under numerous cases in the presence of clock jumps are demonstrated.

• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.1011-1016
• /
• 2020

We propose a framework called Stacked Gated Recurrent Unit - Infrequent Residual Analysis (SG-IRA) that detects anomalies in time-series data that can be trained on streams of raw sensor data without any pre-labeled dataset. To enable such unsupervised learning, SG-IRA includes an estimation model that uses a stacked Gated Recurrent Unit (GRU) structure and an analysis method that detects anomalies based on the difference between the estimated value and the actual measurement (residual). SG-IRA's residual analysis method dynamically adapts the detection threshold from the population using frequency analysis, unlike the baseline model that relies on a constant threshold. In this paper, SG-IRA is evaluated using the industrial control systems (ICS) datasets. SG-IRA improves the detection performance (F1 score) by 5.9% compared to the baseline model.

• Nuclear Engineering and Technology
• /
• v.56 no.4
• /
• pp.1501-1512
• /
• 2024

Plasma disruption in tokamak experiments is a challenging issue that causes damage to the device. Reliable prediction methods are needed, but the lack of full understanding of plasma disruption limits the effectiveness of physics-driven methods. Data-driven methods based on supervised learning are commonly used, and they rely on labelled training data. However, manual labelling of disruption precursors is a time-consuming and challenging task, as some precursors are difficult to accurately identify. The mainstream labelling methods assume that the precursor onset occurs at a fixed time before disruption, which leads to mislabeled samples and suboptimal prediction performance. In this paper, we present disruption prediction methods based on anomaly detection to address these issues, demonstrating good prediction performance on J-TEXT and EAST. By evaluating precursor onset times using different anomaly detection algorithms, it is found that labelling methods can be improved since the onset times of different shots are not necessarily the same. The study optimizes precursor labelling using the onset times inferred by the anomaly detection predictor and test the optimized labels on supervised learning disruption predictors. The results on J-TEXT and EAST show that the models trained on the optimized labels outperform those trained on fixed onset time labels.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.16 no.1
• /
• pp.115-122
• /
• 2006

The most methods for intrusion detection are based on the misuse detection which accumulates hewn intrusion information and makes a decision of an attack against any behavior data. However it is very difficult to detect a new or modified aoack with only the collected patterns of attack behaviors. Therefore, if considering that the method of anomaly behavior detection actually has a high false detection rate, a new approach is required for very huge intrusion patterns based on sequence. The approach can improve a possibility for intrusion detection of known attacks as well as modified and unknown attacks in addition to the similarity measurement of intrusion patterns. This paper proposes a method which applies the multiple sequence alignments technique to the similarity matching of the sequence based intrusion patterns. It enables the statistical analysis of sequence patterns and can be implemented easily. Also, the method reduces the number of detection alerts and false detection for attacks according to the changes of a sequence size.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2024.01a
• /
• pp.45-46
• /
• 2024

본 논문에서는 Self-Attention 기반 딥러닝 기법인 Dense Prediction Transformer(DPT) 모델을 MVTec Anomaly Detection(MVTec AD) 데이터셋에 적용하여 실제 산업 제품 이미지 내 이상 부분을 분할하는 연구를 진행하였다. DPT 모델의 적용을 통해 기존 Convolutional Neural Network(CNN) 기반 이상 탐지기법의 한계점인 지역적 Feature 추출 및 고정된 수용영역으로 인한 문제를 개선하였으며, 실제 산업 제품 데이터에서의 이상 분할 시 기존 주력 기법인 U-Net의 구조를 적용한 최고 성능의 모델보다 1.14%만큼의 성능 향상을 보임에 따라 Self-Attention 기반 딥러닝 기법의 적용이 산업 제품 이상 분할에 효과적임을 입증하였다.