• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.4
• /
• pp.222-230
• /
• 2016

HEMIRE is a 6,000-meter-class remotely operated vehicle (ROV) that has been developed for observation and sampling of objects of interest on the deep seabed. We first carried out deep-seabed exploration around the slopes of the Hupo Bank and the Ulleung Basin in the East Sea in June 2015. Over two weeks, a total of 10 dives were made from a support ship, the R/V Onnuri, at eight stations with water depth ranging between 194 and 2,080 m. The dive times ranged from 1 to 6 hours, depending on the operating conditions. We obtained the following results: 1) video images of the deep seafloor; 2) red snow crab density data (a major fishery resource) and inventories of deep-sea fauna, including an unrecorded organism; 3) specific topographies such as canyons slopes; 4) an undisturbed sediment core obtained using a push corer; and 5) observations of the seabed surface covered with discarded anthropogenic waste material.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.1-16
• /
• 2022

Despite recent breakthroughs in deep learning and computer vision fields, the pixel-wise identification of tiny objects in high-resolution images with complex disturbances remains challenging. This study proposes a modified U-net for tiny crack segmentation in real-world steel-box-girder bridges. The modified U-net adopts the common U-net framework and a novel Self-Attention-Self-Adaption (SASA) neuron as the fundamental computing element. The Self-Attention module applies softmax and gate operations to obtain the attention vector. It enables the neuron to focus on the most significant receptive fields when processing large-scale feature maps. The Self-Adaption module consists of a multiplayer perceptron subnet and achieves deeper feature extraction inside a single neuron. For data augmentation, a grid-based crack random elastic deformation (CRED) algorithm is designed to enrich the diversities and irregular shapes of distributed cracks. Grid-based uniform control nodes are first set on both input images and binary labels, random offsets are then employed on these control nodes, and bilinear interpolation is performed for the rest pixels. The proposed SASA neuron and CRED algorithm are simultaneously deployed to train the modified U-net. 200 raw images with a high resolution of 4928 × 3264 are collected, 160 for training and the rest 40 for the test. 512 × 512 patches are generated from the original images by a sliding window with an overlap of 256 as inputs. Results show that the average IoU between the recognized and ground-truth cracks reaches 0.409, which is 29.8% higher than the regular U-net. A five-fold cross-validation study is performed to verify that the proposed method is robust to different training and test images. Ablation experiments further demonstrate the effectiveness of the proposed SASA neuron and CRED algorithm. Promotions of the average IoU individually utilizing the SASA and CRED module add up to the final promotion of the full model, indicating that the SASA and CRED modules contribute to the different stages of model and data in the training process.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.28 no.4
• /
• pp.372-376
• /
• 2008

We report experimental results on the correlations between welding penetration depth and the frequencies of the radiation from the welding pool. Various welding samples such as SUS304, brass, SUS316, etc. have been investigated with 2 kW CW Nd:YAG laser welding machine. The radiation signals from the plume generated by the interactions between the welding sample and laser with respect to the defocusing length was measured with fiber system collecting the plume signal. Analysis of the frequencies by using fast Fourier transform (FFT) shows that the penetration depth is deep as plume signal frequencies are low, shallow penetration depth for high frequencies. Frequencies up to 250 Hz for obtained signals can be analyzed with the discrete FFT. This is the useful method fur closed loop control of the laser power with respect to the welding penetration depth and is used for real time inspection of the welding quality.

• Journal of Korea Multimedia Society
• /
• v.24 no.7
• /
• pp.880-889
• /
• 2021

In everyday life, recognizing people's emotions from their frames is essential and is a popular research domain in the area of computer vision. Visual emotion has a severe class imbalance in which most of the data are distributed in specific categories. The existing methods do not consider class imbalance and used accuracy as the performance metric, which is not suitable for evaluating the performance of the imbalanced dataset. Therefore, we proposed a method for recognizing visual emotion using balanced data augmentation to address the class imbalance. The proposed method generates a balanced dataset by adopting the random over-sampling and image transformation methods. Also, the proposed method uses the Focal loss as a loss function, which can mitigate the class imbalance by down weighting the well-classified samples. EfficientNet, which is the state-of-the-art method for image classification is used to recognize visual emotion. We compare the performance of the proposed method with that of conventional methods by using a public dataset. The experimental results show that the proposed method increases the F1 score by 40% compared with the method without data augmentation, mitigating class imbalance without loss of classification accuracy.

• Journal of Internet Computing and Services
• /
• v.25 no.2
• /
• pp.21-28
• /
• 2024

Radar is an essential sensor component in autonomous vehicles, and the market for radar applications in this context is steadily expanding with a growing variety of products. In this study, we aimed to enhance the stability and performance of radar systems by developing and evaluating a radar performance prediction model that can predict radar defects. We selected seven machine learning and deep learning algorithms and trained the model with a total of 49 input data types. Ultimately, when we employed an ensemble of 17 models, it exhibited the highest performance. We anticipate that these research findings will assist in predicting product defects at the production stage, thereby maximizing production yield and minimizing the costs associated with defective products.

• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.1
• /
• pp.69-75
• /
• 2023

In recent years, there has been a growing interest in deep learning-based animal pose estimation, especially in the areas of animal behavior analysis and healthcare. However, existing animal pose estimation techniques do not perform well when body parts are occluded or not present. In particular, the occlusion of dog tail or ear might lead to a significant degradation of performance in pet behavior and emotion recognition. In this paper, to solve this intractable problem, we propose a simple yet novel framework for pet pose estimation where pet pose is predicted on an outpainted image where some body parts hidden outside the input image are reconstructed by the image inpainting network preceding the pose estimation network, and we performed a preliminary study to test the feasibility of the proposed approach. We assessed CE-GAN and BAT-Fill for image outpainting, and evaluated SimpleBaseline for pet pose estimation. Our experimental results show that pet pose estimation on outpainted images generated using BAT-Fill outperforms the existing methods of pose estimation on outpainting-less input image.

• Korean Journal of Radiology
• /
• v.22 no.6
• /
• pp.983-993
• /
• 2021

Objective: To investigate the image quality of ultralow-dose CT (ULDCT) of the chest reconstructed using a cycle-consistent generative adversarial network (CycleGAN)-based deep learning method in the evaluation of pulmonary tuberculosis. Materials and Methods: Between June 2019 and November 2019, 103 patients (mean age, 40.8 ± 13.6 years; 61 men and 42 women) with pulmonary tuberculosis were prospectively enrolled to undergo standard-dose CT (120 kVp with automated exposure control), followed immediately by ULDCT (80 kVp and 10 mAs). The images of the two successive scans were used to train the CycleGAN framework for image-to-image translation. The denoising efficacy of the CycleGAN algorithm was compared with that of hybrid and model-based iterative reconstruction. Repeated-measures analysis of variance and Wilcoxon signed-rank test were performed to compare the objective measurements and the subjective image quality scores, respectively. Results: With the optimized CycleGAN denoising model, using the ULDCT images as input, the peak signal-to-noise ratio and structural similarity index improved by 2.0 dB and 0.21, respectively. The CycleGAN-generated denoised ULDCT images typically provided satisfactory image quality for optimal visibility of anatomic structures and pathological findings, with a lower level of image noise (mean ± standard deviation [SD], 19.5 ± 3.0 Hounsfield unit [HU]) than that of the hybrid (66.3 ± 10.5 HU, p < 0.001) and a similar noise level to model-based iterative reconstruction (19.6 ± 2.6 HU, p > 0.908). The CycleGAN-generated images showed the highest contrast-to-noise ratios for the pulmonary lesions, followed by the model-based and hybrid iterative reconstruction. The mean effective radiation dose of ULDCT was 0.12 mSv with a mean 93.9% reduction compared to standard-dose CT. Conclusion: The optimized CycleGAN technique may allow the synthesis of diagnostically acceptable images from ULDCT of the chest for the evaluation of pulmonary tuberculosis.

• The Journal of the Korea Contents Association
• /
• v.22 no.3
• /
• pp.71-80
• /
• 2022

In the past, fish diseases were bacterial in aqua farms, but in recent years, the frequency of fish diseases has increased as they have become viral and mixed. Viral diseases in an enclosed space called a aqua farm have a high spread rate, so it is very likely to lead to mass death. Fast identification of fish diseases is important to prevent group death. However, diagnosis of fish diseases requires a high level of expertise and it is difficult to visually check the condition of fish every time. In order to prevent the spread of the disease, an automatic identification system of diseases or fish is needed. In this paper, in order to improve the performance of the disease identification system of Paralichthys olivaceus based on deep learning, the existing pre-processing method is compared and tested. Target diseases were selected from three most frequent diseases such as Scutica, Vibrio, and Lymphocystis in Paralichthys olivaceus. The RGB, HLS, HSV, LAB, LUV, XYZ, and YCRCV were used as image pre-processing methods. As a result of the experiment, HLS was able to get the best results than using general RGB. It is expected that the fish disease identification system can be advanced by improving the recognition rate of diseases in a simple way.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.24 no.6
• /
• pp.411-418
• /
• 2019

Light-triggered thyristors (LTTs) are essential components in high-power applications, such as HVDC transmission and several pulsed-power applications. Generally, LTT fabrication includes a deep diffusion of aluminum as a p-type dopant to form a uniform p-base region, which needs careful concern for contamination and additional facilities in silicon semiconductor manufacturing factories. We fabricated 4-inch 5,000 V LTTs with boron implantation and diffusion process as a p-type dopant. The LTT contains a main cathode region, edge termination designed with a variation of lateral doping, breakover diode, integrated resistor, photosensitive area, and dV/dt protection region. The doping concentration of each region was adjusted with different doses of boron ion implantation. The fabricated LTTs showed good light triggering characteristics for a light pulse of 905 nm and a blocking voltage (V_DRM) of 6,500 V. They drove an average on-state current (I_TAVM) of 2,270 A, peak nonrepetitive surge current (I_TSM) of 61 kA, critical rate of rise of on-state current (di/dt) of 1,010 A/㎲, and limiting load integral (I²T) of 17 MA²s without damage to the device.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.4103-4103
• /
• 2001

The aim of this study was to examine whether near infrared spectroscopy (NIRS) is an acceptable tool to determine cholesterol and collagen in human atherosclerotic plaque without destruction of the analyzed areas and without danger the endothelial cells - three preconditions for the development of a NIR-heart-catheter. The questions were: Can the cholesterol and collagen content of the arterial intima be estimated with acceptable precision in vitro by NIRS despite the matrix inhomogeneity of the plaques and their anatomic variability\ulcorner How deep can such NIR radiation penetrate into arterial tissue without danger for endothelial cells\ulcorner Is this penetration sufficient for information on the lipid and collagen accumulation\ulcorner Using NIRS, cholesterol and collagen can be determined with acceptable precision in model mixtures and human aortic specimens (r=0,896 to 0,957). The chemical reference method was HPLC. The energy dose was 71 mW/$cm^{-2}$ using a fiber optic strand with a length of 1.5m and an optical window of d=4mm. This dose appears to be not dangerous for endothelial cells, It will be attenuated to 50% by a arterial tissue of about 170-$200\mu\textrm{m}$ thickness. The results are also acceptable using a thin coronary catheter-like fiber optic strand (d=1mm).