• Journal of the Korea Institute of Military Science and Technology
• /
• v.26 no.3
• /
• pp.234-245
• /
• 2023

The Arena Fragmentation Test(AFT) is designed to analyze warhead performance by measuring fragmentation data. In order to evaluate the results of the AFT, a set of AFT images are captured by high-speed cameras. To detect objects in the AFT image set, ResNet-50 based Faster R-CNN is used as a detection model. However, because of the low resolution of the AFT image set, a detection model has shown low performance. To enhance the performance of the detection model, Super-resolution(SR) methods are used to increase the AFT image set resolution. To this end, The Bicubic method and three SR models: ZSSR, EDSR, and SwinIR are used. The use of SR images results in an increase in the performance of the detection model. While the increase in the number of pixels representing a fragment flame in the AFT images improves the Recall performance of the detection model, the number of pixels representing noise also increases, leading to a slight decreases in Precision performance. Consequently, the F1 score is increased by up to 9 %, demonstrating the effectiveness of SR in enhancing the performance of the detection model.

• The Journal of the Acoustical Society of Korea
• /
• v.43 no.3
• /
• pp.305-313
• /
• 2024

For effective bi-static sonar operation, detection performance analysis must be performed reflecting the characteristics of sound propagation due to the ocean environment and target information. However, previous studies analyzing bistatic sonar detection performance have either not considered the ocean environment and target characteristics or have been conducted using simplified approaches. Therefore, in this study, we compared and analyzed the bistatic detection performance in Yellow sea and Ulleung basin both with and without considering target characteristics. A numerical analysis model was used to derive an accurate bistatic target strength for the submarine-shaped target, and signal excess was calculated by reflecting the simulated target strength. As a result, significant changes in detection performance were observed depending on the source and receiver locations as well as the target strength.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.19-21
• /
• 2022

Anomaly detection using self-supervised learning typically generates synthetic data to learn to classify normal and abnormal, and uses real abnormal data as test data to measure anomaly detection performance. In a study using this method to generate synthetic data similar to normal data, anomaly detection was carried out by generating synthetic data by cutting and pasting a specific patch from the original image. In this way, the degree of similarity to normal data depends on the number and size of patches, which affects anomaly detection performance. In this paper, synthetic data were generated by varying patch sizes and numbers, and then similarity and analysis with normal data were conducted using a pre-trained model, and anomaly detection performance was measured by learning the model.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.67 no.1
• /
• pp.27-32
• /
• 2018

In this paper, we developed an algorithm to detect arc of PV power plant through frequency analysis. For arc detection based on frequency analysis, the filter should be designed to emphasize the difference between the arc state and the normal state. Therefore, in this paper, we analyzed the arc detection performance according to various filter structures. The arc detection algorithm developed in this paper extracts the filtering signal on current by using various filters and then calculates the frequency components and total energy using the FFT. In the final step, the arc is detected using the calculated energy magnitude. In order to verify the performance of the proposed arc detection algorithm, experiments were conducted on 51 kW solar inverters connected to power line. Through various experiments, it was confirmed that the proposed method effectively detects the arc.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.9 no.1 s.24
• /
• pp.24-32
• /
• 2006

This paper is to analyze the effectiveness of multistatic sonar network based on detection performance. The multistatic sonar network is a distributed detection system that places a source and multi-receivers apart. So it needs a detection technique that relates to decision rule and optimization of sonar system to improve the detection performance. For this we propose a data fusion procedure using Bayesian decision and optimal sensor arrangement by optimizing a bistatic sonar. Also, to analyze the detection performance effectively, we propose the environmental model that simulates a propagation loss and target strength suitable for multistatic sonar networks in real surroundings. The effectiveness analysis on the multistatic sonar network confirms itself as a promising tool for effective allocation of detection resources in multistatic sonar system.

• Journal of Digital Contents Society
• /
• v.19 no.4
• /
• pp.711-717
• /
• 2018

The breakthrough in computer and network has facilitated a variety of information exchange. However, at the same time, malicious users and groups are attacking vulnerable systems. Intrusion Detection System(IDS) detects malicious behaviors through network packet analysis. However, it has a burden of processing a large amount of packets in a short time. Therefore, in order to solve these problem, we propose a network intrusion detection system that operates at kernel level to improve detection performance at user level. In fact, we confirmed that the network intrusion detection system implemented at kernel level improves packet analysis and detection performance.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.5
• /
• pp.225-231
• /
• 2023

With the recent surge in the autonomous driving market, the significance of lane detection technology has escalated. Lane detection plays a pivotal role in autonomous driving systems by identifying lanes to ensure safe vehicle operation. Traditional lane detection models rely on engineers manually extracting lane features from predefined environments. However, real-world road conditions present diverse challenges, hampering the engineers' ability to extract adaptable lane features, resulting in limited performance. Consequently, recent research has focused on developing deep learning based lane detection models to extract lane features directly from data. In this paper, we classify lane detection models into four categories: cluster-based, curve-based, information propagation-based, and anchor-based methods. We conduct an extensive analysis of the strengths and weaknesses of each approach, evaluate the model's performance on an embedded board, and assess their practicality and effectiveness. Based on our findings, we propose future research directions and potential enhancements.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.2
• /
• pp.61-66
• /
• 2005

In this paper, we analyzed the error rate Performance of Convolution coded 16 QAM signal with Optimum Threshold Detection in Rician Fading Enviroments. The performance of 16-QAM signal with CTD (conventional threshold detection) which employs convolution coding technique was analyzed and the performance improvement of convolution coded 16-QAM signal with OTD (optimum threshold detection) which is varied according to fading parameter 'K' and AWGN in Rician Fading channel was simulated. As a result of analysis, it was shown the effect of performance improvement to overcome the environment of mobile radio data communication channel.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.26 no.4
• /
• pp.271-276
• /
• 2021

Arc fault detector based on multilevel DWT with analysis of high-frequency noise components over 100 kHz is proposed in this study to improve the performance in detecting serial arcs and distinguishing them from inverter noise in PV systems. PV inverters generally operate at a frequency range of 20-50 kHz for switching operation and maximum power tracking control, and the effect of these frequency components on the signal for arc detection leads to negative arc detection. High-speed ADC and multilevel DWT are used in this study to analyze frequency components above 100 kHz. Such high frequency components are less influenced by inverter noise and utilized to detect as well as separate DC series arc from inverter noise. Arc detectors identify the input current of PV inverters using a Rogowski coil. The sensed signal is filtered, amplified, and used in 800kSPS ADC and DWT analysis and arc occurrence determination in DSP. An arc detection simulation facility in UL1699B was constructed and AFD tests the proposed detector were conducted to verify the performance of arc detection and performance of distinction of the negative arc. The satisfactory performance of the arc detector meets the standard of arc detection and extinguishing time of UL1699B with an arc detection time of approximately 0.11 seconds.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.19 no.4
• /
• pp.117-126
• /
• 2023

Traditional markets, formed naturally, often feature aged buildings and facilities that are susceptible to fire. However, the lack of adequate fire detection systems in these markets can easily lead to large-scale fires upon ignition. Therefore, this study was conducted with the aim of detecting fires in traditional markets, utilizing thermal imaging cameras for data collection and the YOLOv8 model for object detection experiments. Data were collected in the night markets within traditional markets of xx city and by simulating fire scenarios. A comparative analysis of the Nano and XL models of YOLOv8 revealed that the XL model is more effective in detecting fires. The XL model not only demonstrated higher accuracy in correctly identifying flames but also tended to miss fewer fires compared to the Nano model. In the case of objects other than flames, the XL model showed superior performance over the Nano model. Taking all these factors into account, it is anticipated that with further data collection and improvement in model performance, a suitable fire detection system for traditional markets can be developed.