JCJungyu Choi

Jungyu Choi

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Jungyu Choi has a total of 10 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Multi-Task U-Net with Class-Conditional Mahalanobis Scoring for Unsupervised Drone Anomaly Detection

    by Jeongin Ju, Jungyu Choi, Seowoo Kim, Sungbin Im

    With the growing use of drones, the importance of early detection of abnormal conditions has increased. However, in real-world scenarios, it is difficult to collect sufficient abnormal data in advance, making unsupervised anomaly detection approaches essential. Conventional autoencoder (AE)-based methods typically use reconstruction error as the anomaly score, but their detection performance degrades when abnormal samples are reconstructed as normal. To address this limitation, this paper proposes a multi-task U-Net that simultaneously performs reconstruction and classification. By applying structured loss functions—Supervised Contrastive (SupCon), Triplet, and Center losses—the proposed model shapes the latent representation into class-conditional Gaussian distributions and computes anomaly scores based on the Mahalanobis distance. Furthermore, the integration of Squeeze-and-Excitation (SE) blocks and an adaptive noise layer enables robust representation learning even under low signal-to-noise ratio (SNR) conditions. Experimental results on the ICSV31-AI Challenge drone acoustic dataset demonstrate that the proposed model achieved a macro Area Under the Curve (AUC) of 0.9630, representing over a 59\% relative improvement compared to the baseline WaveNet AE (0.6059) and significantly outperforming the single-task U-Net (0.5620). These results confirm that the proposed method enables precise, class-aware anomaly detection under unsupervised settings and holds strong potential for applications beyond drones, including various types of industrial equipment.

  • Open Access

    DUAL-RES: A Phase-Aware Lightweight Framework for Real-Time Denoising and Emergency-Caller Detection

    by Joonhwi Kim, Jungyu Choi, Sungbin Im

    Reliable acoustic instrumentation and measurement in disaster environments is challenging because structure-borne signals from survivors are often masked by highly nonstationary noise from machinery, sirens, and impacts. In particular, phase distortion severely degrades the accuracy of time-difference-of-arrival (TDOA) measurements used for source localization. To enhance measurement robustness, this paper introduces Dual-objective Ultra-lightweight Audio Low-power — Rescue-recognition & Enhancement System (DUAL–RES), a lightweight, phase-aware framework for real-time acoustic denoising and event detection. The system operates in the complex short-time Fourier transform (STFT) domain, estimating gain masks for both magnitude and phase components and applying waveform-domain refinement, while an integrated convolutional classifier enables simultaneous detection and characterization of structurally relevant acoustic events. Comprehensive evaluation across three datasets shows that DUAL–RES improves measurement reliability under noise: (1) on VoiceBank + DEMAND, it achieves competitive perceptual metrics with a compact 0.81M-parameter profile; (2) in controlled simulations, it increases classification accuracy at low signal-to-noise ratios (SNRs) and reduces phase distortion relative to representative baselines; and (3) on real microphone-array recordings, it preserves localization accuracy close to the unprocessed mixture, avoiding the degradation commonly observed with other denoisers. With real-time inference on embedded platforms such as the Jetson Nano, DUAL–RES provides a practical instrumentation framework for deployable acoustic measurement, monitoring, and event-based detection in disaster scenarios.


  • Open Access

    Multivariate Kernel Density Estimation Performance in Detecting Abnormalities in Three-Phase Induction Motors With a Focus on Optimizing Kernel and Bandwidth

    by Jungyu Choi, Sungbin Im

    In this study, we focus on the application of anomaly detection for three-phase induction motors using current data, a crucial aspect of predictive maintenance technologies transforming industrial operations. Induction motors play a pivotal role in converting electrical energy to mechanical energy across various industries. Detecting anomalies in these motors is essential for ensuring operational efficiency and minimizing downtime. Our approach involves analyzing the spectral characteristics of current data to identify abnormal motor states. Specifically, we leverage the distortion observed in the current spectrum as a key feature indicative of motor anomalies. To achieve this, we employ a multivariate kernel density estimation (MKDE) algorithm, which operates under unsupervised learning principles for anomaly detection. To determine the optimal threshold for anomaly detection, we utilize the Youden index, enhancing the reliability of our detection method. Furthermore, we fine-tune the model’s hyperparameters, including the kernel bandwidth, through experimentation with various selection rules. The final bandwidth is chosen based on an analysis of the feature distribution within the data. Implementing our proposed method offers significant benefits to industries by potentially reducing maintenance costs through timely anomaly detection in induction motors. Proactively replacing motor components when anomalies are detected can enhance overall motor performance and extend their operational lifespan. This approach underscores the critical role of advanced anomaly detection techniques in improving industrial efficiency and reliability.


  • Open Access

    Application of CNN Models to Detect and Classify Leakages in Water Pipelines Using Magnitude Spectra of Vibration Sound

    by Jungyu Choi, Sungbin Im

    Conventional schemes to detect leakage in water pipes require leakage exploration experts. However, to save time and cost, demand for sensor-based leakage detection and automated classification systems is increasing. Therefore, in this study, we propose a convolutional neural network (CNN) model to detect and classify water leakage using vibration data collected by leakage detection sensors installed in water pipes. Experiment results show that the proposed CNN model achieves an F1-score of 94.82% and Matthew’s correlation coefficient of 94.47%, whereas the corresponding values for a support vector machine model are 80.99% and 79.86%, respectively. This study demonstrates the superior performance of the CNN-based leakage detection scheme with vibration sensors. This can help one to save detection time and cost incurred by skilled engineers. In addition, it is possible to develop an intelligent leak detection system based on the proposed one.


  • Open Access

    Supervised Multivariate Kernel Density Estimation for Enhanced Plasma Etching Endpoint Detection

    by Jungyu Choi, Bobae Kim, Sungbin Im, Geonwook Yoo

    The advancement of semiconductor technology nodes requires precise control of their manufacturing process, including plasma etching, which is highly important in terms of the yield, cost, and device performance. Endpoint detection (EPD) is an imperative technique for controlling this process. Here, we propose a novel EPD scheme based on multivariate kernel density estimation (MKDE). The proposed approach is developed by extending the conventional unsupervised learning MKDE method to supervised learning. The performance of the proposed scheme is validated on randomly selected optical emission spectroscopy data collected from an industrial semiconductor manufacturing process. Because the proposed approach uses target values (labeling) of data, it demonstrates enhanced EPD performance compared to the conventional MKDE method, even without threshold presetting.


  • EVM Performance Comparison of a Repeater under Nonlinear Interference of Passive Devices

    by Jungyu Choi, Zhihui Jin, Jinzi Song, Sungbin Im

    본 논문에서, FDD (Frequency Division Duplex) 무선 통신 시스템에서 DL (Downlink) 신호로 인한 3차 수동소자 상호변조 왜곡 (PIMD)이 중계기의 EVM 성능에 미치는 영향을 비교한다. PIMD는 무선 통신 환경에서 수동소자의 비선형성으로 인해 발생하는 간섭신호이며, 수신기의 잡음수준을 높여 통신의 품질을 저하시킨다. 따라서 이 실험에서는 LTE UL (Uplink) 신호를 생성 후 QPSK, 16QAM 방식에서 Gaussian Noise와 PIMD의 간섭이 있을 때, SNR, SIR에 따른 EVM을 측정한다. 그리고 실험 결과를 통해 PIMD 간섭이 있을 시 중계기의 성능 저하 현상을 살펴본다. 본 연구의 결과를 이용하여 PIMD 제거기의 성능에 따른 개선범위를 도출할 수 있다.