• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1109-1124
• /
• 2022

This study proposes a marine debris monitoring methods using satellite and drone multispectral images. A multi-layer perceptron (MLP) model was applied to detect marine debris using Sentinel-2 satellite image. And for the detection of marine debris using drone multispectral images, performance evaluation and comparison of U-Net, DeepLabv3+ (ResNet50) and DeepLabv3+ (Inceptionv3) among deep learning models were performed (mIoU 0.68). As a result of marine debris detection using satellite image, the F1-Score was 0.97. Marine debris detection using drone multispectral images was performed on vegetative debris and plastics. As a result of detection, when DeepLabv3+ (Inceptionv3) was used, the most model accuracy, mean intersection over union (mIoU), was 0.68. Vegetative debris showed an F1-Score of 0.93 and IoU of 0.86, while plastics showed low performance with an F1-Score of 0.5 and IoU of 0.33. However, the F1-Score of the spectral index applied to generate plastic mask images was 0.81, which was higher than the plastics detection performance of DeepLabv3+ (Inceptionv3), and it was confirmed that plastics monitoring using the spectral index was possible. The marine debris monitoring technique proposed in this study can be used to establish a plan for marine debris collection and treatment as well as to provide quantitative data on marine debris generation.

• Tunnel and Underground Space
• /
• v.34 no.3
• /
• pp.208-217
• /
• 2024

There is so much research effort for developing and implementing deep learning-based surveillance systems to manage health and safety issues in construction sites. Especially, the development of deep learning-based object detection in various environmental changes has been progressing because those affect decreasing searching performance of the model. Among the various environmental variables, the accuracy of the object detection model is significantly dropped under low illuminance, and consistent object detection accuracy cannot be secured even the model is trained using low-light images. Accordingly, there is a need of low-light enhancement to keep the performance under low illuminance. Therefore, this paper conducts a comparative study of various deep learning-based low-light image enhancement models (GLADNet, KinD, LLFlow, Zero-DCE) using the acquired construction site image data. The low-light enhanced image was visually verified, and it was quantitatively analyzed by adopting image quality evaluation metrics such as PSNR, SSIM, Delta-E. As a result of the experiment, the low-light image enhancement performance of GLADNet showed excellent results in quantitative and qualitative evaluation, and it was analyzed to be suitable as a low-light image enhancement model. If the low-light image enhancement technique is applied as an image preprocessing to the deep learning-based object detection model in the future, it is expected to secure consistent object detection performance in a low-light environment.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.14 no.5
• /
• pp.237-242
• /
• 2014

In this paper, we find out the lifetime index in order to determine the time-dependent trend of deteriorating performance of 6.6kV high-voltage power cable in operation at a power station. The cable systems used in our study have been in operation for 13 years. With measurements for the 13 years, we analyzed the insulation resistances. By developing measuring equipment (comprized mainly of transformer, temperature sensor, and LPF) operating by the three-phase electric power, we analyzed the changing characteristics of insulation resistance of power cable. In contrast to 22kV cables, 6.6 kV cables have thicker insulation. Therefore the characteristics of 6.6kV cables are different from that of 22kV cables. The study found that as time passes, the insulation resistance does not decrease continuously; it decreases to a certain value, then does not decrease any more and shows properties of oscillation. We could not detect the process of deterioration in the preceding twelve years. The cable system showed great stability so that deterioration was not apparent. In this case, it is not possible to measure the future life indices of power cables because the lifetime indices are not predictable

• Journal of the Korean Society of Propulsion Engineers
• /
• v.23 no.4
• /
• pp.21-27
• /
• 2019

Composite motor cases fabricated by the filament winding method are structurally weak in the dome when they are required to withstand the internal pressure of the combustion gas. In this study, a finite element analysis is conducted to compare the burst pressure of a composite dome according to the variation of the pressure distribution ratio(PDR). The performance of the composite motor case was compared quantitatively by calculating the stress on the inner and outer dome surfaces and metal boss volume. As a result, the critical point of the failure mode was observed at a PDR between 2.5 and 3.0. A design at a PDR of 2.5­-3.5 can reduce the weight of metal boss without fluctuation in the burst pressure of the combustion motor case. Moreover as the design reference value changes according to the dome shape and opening size, further analysis and testing are necessary.

• Geophysics and Geophysical Exploration
• /
• v.25 no.3
• /
• pp.140-161
• /
• 2022

Seismic data with missing traces are often obtained regularly or irregularly due to environmental and economic constraints in their acquisition. Accordingly, seismic data interpolation is an essential step in seismic data processing. Recently, research activity on machine learning-based seismic data interpolation has been flourishing. In particular, convolutional neural network (CNN) and generative adversarial network (GAN), which are widely used algorithms for super-resolution problem solving in the image processing field, are also used for seismic data interpolation. In this study, CNN-based algorithm, U-Net and GAN-based algorithm, and conditional Wasserstein GAN (cWGAN) were used as seismic data interpolation methods. The results and performances of the methods were evaluated thoroughly to find an optimal interpolation method, which reconstructs with high accuracy missing seismic data. The work process for model training and performance evaluation was divided into two cases (i.e., Cases I and II). In Case I, we trained the model using only the regularly sampled data with 50% missing traces. We evaluated the model performance by applying the trained model to a total of six different test datasets, which consisted of a combination of regular, irregular, and sampling ratios. In Case II, six different models were generated using the training datasets sampled in the same way as the six test datasets. The models were applied to the same test datasets used in Case I to compare the results. We found that cWGAN showed better prediction performance than U-Net with higher PSNR and SSIM. However, cWGAN generated additional noise to the prediction results; thus, an ensemble technique was performed to remove the noise and improve the accuracy. The cWGAN ensemble model removed successfully the noise and showed improved PSNR and SSIM compared with existing individual models.

• Korean Journal of Environment and Ecology
• /
• v.38 no.2
• /
• pp.178-188
• /
• 2024

The dynamics of suspended sediment (SS) in forested catchments vary depending upon human or natural disturbances, including land use change, forestry activity, forest fires, and landslides. Understanding the dynamics of SS originating from the potential sources within a forested catchment is crucial for establishing an effective water quality management strategy. Therefore, to suggest a systematic method for interpreting SS dynamics, we evaluated the performance and applicability of ten methods for calculating the hysteresis index based on observed hydrological data and two calculation models (Lawler's method and Lloyd's method) with five sampling intervals (50th, 25th, 10th, 5th, and 1st percentiles). Our results showed that Lloyd's method, which used a sampling interval at the 1st percentile, had the largest number of analyzable runoff events and exhibited the best performance. The results of this study can contribute to quantifying the hysteresis in the relationship between discharge and SS and provide useful information for interpreting SS dynamics.

• Korean Journal of Remote Sensing
• /
• v.37 no.3
• /
• pp.475-491
• /
• 2021

With the recent increase in available high-resolution (< ~1 m) satellite SAR images, the demand for precise registration of SAR images is increasing in various fields including change detection. The registration between high-resolution SAR images acquired in different look angle is difficult due to speckle noise and geometric distortion caused by the characteristics of SAR images. In this study, registration is performed in two stages, coarse and fine, using the x-band SAR data imaged at staring spotlight mode of TerraSAR-X. For the coarse registration, a method combining the adaptive sampling method and SAR-SIFT (Scale Invariant Feature Transform) is applied, and three rigid methods (NCC: Normalized Cross Correlation, Phase Congruency-NCC, MI: Mutual Information) and one non-rigid (Gefolki: Geoscience extended Flow Optical Flow Lucas-Kanade Iterative), for the fine registration stage, was performed for performance comparison. The results were compared by using RMSE (Root Mean Square Error) and FSIM (Feature Similarity) index, and all rigid models showed poor results in all image combinations. It is confirmed that the rigid models have a large registration error in the rugged terrain area. As a result of applying the Gefolki algorithm, it was confirmed that the RMSE of Gefolki showed the best result as a 1~3 pixels, and the FSIM index also obtained a higher value than 0.02~0.03 compared to other rigid methods. It was confirmed that the mis-registration due to terrain effect could be sufficiently reduced by the Gefolki algorithm.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2011.04a
• /
• pp.315-318
• /
• 2011

본 논문에서는 구조물 연결부의 실시간 손상 검색을 통해 이상이 감지되었을 경우 자가치유까지 가능한 지능형 볼트접합부 시스템에 관한 실험적 연구결과가 제시되었다. 지능형 센서인 PZT센서의 전기-역학적 커플링 특성을 이용한 전기역학적 임피던스 기반의 구조물 건전성 평가 방법이 사용되었다. 전기역학적 임피던스의 측정을 통한 계측값을 베이스라인 값과 비교하는 손상 평가를 통해 구조물 볼트접합부의 볼트풀림 손상을 진단하고, 손상은 손상지수 RMSD를 통하여 정량화되었다. 볼트접합부의 손상이 감지되었을 경우 형상기억합금(SMA) 와셔에 부착되어있는 히팅 필름에 전원을 가함으로써 형상기억합금에 열을 가하고, 가열된 형상기억합금 와셔는 축방향으로 팽창을 함으로써 잃었던 볼트의 토크력을 회복시켜주었다. 실험 결과, 제안된 전기역학적 임피던스 기반의 구조물 건전성 평가기법과 형상기억합금 와셔 기반의 볼트접합부 자가치유 시스템의 성능 평가와 검증이 이루어졌다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.3
• /
• pp.103-108
• /
• 2016

In daily life, humans get stress very often. Stress is one of the important factors of healthy life and closely related to the quality of life. Too much stress is known to cause hormone imbalance of our body, and it is observed by the brain and bio signals. Based on this, the relationship between brain signal and stress is explored, and brain signal based stress index is proposed in our work. In this study, an EEG measurement device with 32 channels is adopted. However, only two channels (FP1, FP2) are used to this study considering the applicability of the proposed method in real enveironment, and to compare it with the commercial 2 channel EEG device. Frequency domain features are power of each frequency bands, subtraction, addition, or division by each frequency bands. Features in time domain are hurst exponent, correlation dimension, lyapunov exponent, etc. Total 6 subjects are participated in this experiment with English sentence reading task given. Among several candidate features, ${\frac{{\theta}\;power}{mid\;{\beta}\;power}}$ shows the best test performance (70.8%). For future work, we will confirm the results is consistent in low price EEG device.

• Journal of the Korean Data and Information Science Society
• /
• v.8 no.1
• /
• pp.21-30
• /
• 1997

In accelerated life tests, the failure time of an item is observed under a high stress level, and based on the time the performances of items are investigated at the normal stress level. In this paper, when the mean of the prior of a failure rate is known in the exponential lifetime distribution with censored accelerated failure time data, we utilize the empirical Bayesian method by using the moment estimators in order to estimate the parameters of the prior distribution and obtain the empirical Bayesian predictive density and predictive intervals for a future observation under the normal stress level.