• Korean Journal of Radiology
• /
• v.24 no.4
• /
• pp.294-304
• /
• 2023

Objective: We aimed to investigate whether image standardization using deep learning-based computed tomography (CT) image conversion would improve the performance of deep learning-based automated hepatic segmentation across various reconstruction methods. Materials and Methods: We collected contrast-enhanced dual-energy CT of the abdomen that was obtained using various reconstruction methods, including filtered back projection, iterative reconstruction, optimum contrast, and monoenergetic images with 40, 60, and 80 keV. A deep learning based image conversion algorithm was developed to standardize the CT images using 142 CT examinations (128 for training and 14 for tuning). A separate set of 43 CT examinations from 42 patients (mean age, 10.1 years) was used as the test data. A commercial software program (MEDIP PRO v2.0.0.0, MEDICALIP Co. Ltd.) based on 2D U-NET was used to create liver segmentation masks with liver volume. The original 80 keV images were used as the ground truth. We used the paired t-test to compare the segmentation performance in the Dice similarity coefficient (DSC) and difference ratio of the liver volume relative to the ground truth volume before and after image standardization. The concordance correlation coefficient (CCC) was used to assess the agreement between the segmented liver volume and ground-truth volume. Results: The original CT images showed variable and poor segmentation performances. The standardized images achieved significantly higher DSCs for liver segmentation than the original images (DSC [original, 5.40%-91.27%] vs. [standardized, 93.16%-96.74%], all P < 0.001). The difference ratio of liver volume also decreased significantly after image conversion (original, 9.84%-91.37% vs. standardized, 1.99%-4.41%). In all protocols, CCCs improved after image conversion (original, -0.006-0.964 vs. standardized, 0.990-0.998). Conclusion: Deep learning-based CT image standardization can improve the performance of automated hepatic segmentation using CT images reconstructed using various methods. Deep learning-based CT image conversion may have the potential to improve the generalizability of the segmentation network.

• Imaging Science in Dentistry
• /
• v.54 no.1
• /
• pp.81-91
• /
• 2024

Purpose: The objective of this study was to propose a deep-learning model for the detection of the mandibular canal on dental panoramic radiographs. Materials and Methods: A total of 2,100 panoramic radiographs (PANs) were collected from 3 different machines: RAYSCAN Alpha (n=700, PAN A), OP-100 (n=700, PAN B), and CS8100 (n=700, PAN C). Initially, an oral and maxillofacial radiologist coarsely annotated the mandibular canals. For deep learning analysis, convolutional neural networks (CNNs) utilizing U-Net architecture were employed for automated canal segmentation. Seven independent networks were trained using training sets representing all possible combinations of the 3 groups. These networks were then assessed using a hold-out test dataset. Results: Among the 7 networks evaluated, the network trained with all 3 available groups achieved an average precision of 90.6%, a recall of 87.4%, and a Dice similarity coefficient (DSC) of 88.9%. The 3 networks trained using each of the 3 possible 2-group combinations also demonstrated reliable performance for mandibular canal segmentation, as follows: 1) PAN A and B exhibited a mean DSC of 87.9%, 2) PAN A and C displayed a mean DSC of 87.8%, and 3) PAN B and C demonstrated a mean DSC of 88.4%. Conclusion: This multi-device study indicated that the examined CNN-based deep learning approach can achieve excellent canal segmentation performance, with a DSC exceeding 88%. Furthermore, the study highlighted the importance of considering the characteristics of panoramic radiographs when developing a robust deep-learning network, rather than depending solely on the size of the dataset.

• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.335-349
• /
• 2023

Bridges constantly undergo deterioration and damage, the most common ones being concrete damage and exposed rebar. Periodic inspection of bridges to identify damages can aid in their quick remediation. Likewise, identifying components can provide context for damage assessment and help gauge a bridge's state of interaction with its surroundings. Current inspection techniques rely on manual site visits, which can be time-consuming and costly. More recently, robotic inspection assisted by autonomous data analytics based on Computer Vision (CV) and Artificial Intelligence (AI) has been viewed as a suitable alternative to manual inspection because of its efficiency and accuracy. To aid research in this avenue, this study performs a comparative assessment of different architectures, loss functions, and ensembling strategies for the autonomous segmentation of bridge components and damages. The experiments lead to several interesting discoveries. Nested Reg-UNet architecture is found to outperform five other state-of-the-art architectures in both damage and component segmentation tasks. The architecture is built by combining a Nested UNet style dense configuration with a pretrained RegNet encoder. In terms of the mean Intersection over Union (mIoU) metric, the Nested Reg-UNet architecture provides an improvement of 2.86% on the damage segmentation task and 1.66% on the component segmentation task compared to the state-of-the-art UNet architecture. Furthermore, it is demonstrated that incorporating the Lovasz-Softmax loss function to counter class imbalance can boost performance by 3.44% in the component segmentation task over the most employed alternative, weighted Cross Entropy (wCE). Finally, weighted softmax ensembling is found to be quite effective when used synchronously with the Nested Reg-UNet architecture by providing mIoU improvement of 0.74% in the component segmentation task and 1.14% in the damage segmentation task over a single-architecture baseline. Overall, the best mIoU of 92.50% for the component segmentation task and 84.19% for the damage segmentation task validate the feasibility of these techniques for autonomous bridge component and damage segmentation using RGB images.

• Nuclear Engineering and Technology
• /
• v.56 no.1
• /
• pp.167-179
• /
• 2024

This study explores the feasibility of employing (U, Th)-based accident tolerant fuels (ATFs), specifically (_0.8UO₂, _0.2ThO₂), (_0.8UN, _0.2ThN), and (_0.8UC, _0.2ThC). The investigation assesses the overall performance of these proposed fuel materials in comparison to the conventional UO₂, focusing on deep neutronic and thermal-hydraulic (Th) analyses. Neutronic analysis utilized the MCNPX code, while COMSOL Multiphysics was employed for thermal-hydraulic analysis. The primary objective of this research is to overcome the limitations associated with traditional UO₂ fuel by exploring alternative fuel materials that offer advantages in terms of abundance and potential improvements in performance and safety. Given the limited abundance of UO₂, long-term sustainable nuclear energy production faces challenges. From a neutronic standpoint, the U-Th based fuels demonstrated remarkable fuel cycle lengths, except (_0.8UN, _0.2ThN), which exhibited the minimum fuel cycle length and, consequently, the lowest fuel burn-up. Regarding thermal-hydraulic performance, (_0.8UN, _0.2ThN) exhibited outstanding performance with significant margins against fuel melting compared to the other materials. Overall, when considering the integrated performance, the most favourable results were obtained with the use of the (_0.8UC, _0.2ThC) fuel configurations. This study contributes valuable insights into the potential benefits of (U, Th)-based ATFs as a promising avenue for enhanced nuclear fuel performance.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.64 no.3
• /
• pp.63-73
• /
• 2022

A more accurate understanding of the irrigation water supply is necessary for efficient agricultural water management. Although we measure water levels in an irrigation canal using ultrasonic water level gauges, some errors occur due to malfunctions or the surrounding environment. This study aims to apply CNN (Convolutional Neural Network) Deep-learning-based image classification and segmentation models to the irrigation canal's CCTV (Closed-Circuit Television) images. The CCTV images were acquired from the irrigation canal of the agricultural reservoir in Cheorwon-gun, Gangwon-do. We used the ResNet-50 model for the image classification model and the U-Net model for the image segmentation model. Using the Natural Breaks algorithm, we divided water level data into 2, 4, and 8 groups for image classification models. The classification models of 2, 4, and 8 groups showed the accuracy of 1.000, 0.987, and 0.634, respectively. The image segmentation model showed a Dice score of 0.998 and predicted water levels showed R² of 0.97 and MAE (Mean Absolute Error) of 0.02 m. The image classification models can be applied to the automatic gate-controller at four divisions of water levels. Also, the image segmentation model results can be applied to the alternative measurement for ultrasonic water gauges. We expect that the results of this study can provide a more scientific and efficient approach for agricultural water management.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.24 no.3
• /
• pp.83-98
• /
• 2021

Urban green space is an important component for enhancing urban ecosystem health. Thus, identifying the spatial structure of urban green space is required to manage a healthy urban ecosystem. The Ministry of Environment has provided the level 3 land cover map(the highest (1m) spatial resolution map) with a total of 41 classes since 2010. However, specific urban green information such as street trees was identified just as grassland or even not classified them as a vegetated area in the map. Therefore, this study classified detailed urban green information(i.e., tree, shrub, and grass), not included in the existing level 3 land cover map, using two types of high-resolution(<1m) remote sensing data(i.e., airborne LiDAR and RGB ortho imagery) in Suwon, South Korea. U-Net, one of image segmentation deep learning approaches, was adopted to classify detailed urban green space. A total of three classification models(i.e., LRGB10, LRGB5, and RGB5) were proposed depending on the target number of classes and the types of input data. The average overall accuracies for test sites were 83.40% (LRGB10), 89.44%(LRGB5), and 74.76%(RGB5). Among three models, LRGB5, which uses both airborne LiDAR and RGB ortho imagery with 5 target classes(i.e., tree, shrub, grass, building, and the others), resulted in the best performance. The area ratio of total urban green space(based on trees, shrub, and grass information) for the entire Suwon was 45.61%(LRGB10), 43.47%(LRGB5), and 44.22%(RGB5). All models were able to provide additional 13.40% of urban tree information on average when compared to the existing level 3 land cover map. Moreover, these urban green classification results are expected to be utilized in various urban green studies or decision making processes, as it provides detailed information on urban green space.

• Korean Journal of Remote Sensing
• /
• v.38 no.1
• /
• pp.111-126
• /
• 2022

Land use and land cover (LULC) mapping is an important factor in geospatial analysis. Although highly precise ground-based LULC monitoring is possible, it is time consuming and costly. Conversely, because the synthetic aperture radar (SAR) sensor is an all-weather sensor with high resolution, it could replace field-based LULC monitoring systems with low cost and less time requirement. Thus, LULC is one of the major areas in SAR applications. We developed a LULC model using only KOMPSAT-5 single co-polarized data and digital elevation model (DEM) data. Twelve HH-polarized images and 18 VV-polarized images were collected, and two HH-polarized images and four VV-polarized images were selected for the model testing. To train the LULC model, we applied the conditional generative adversarial network (cGAN) method. We used U-Net combined with the residual unit (ResUNet) model to generate the cGAN method. When analyzing the training history at 1732 epochs, the ResUNet model showed a maximum overall accuracy (OA) of 93.89 and a Kappa coefficient of 0.91. The model exhibited high performance in the test datasets with an OA greater than 90. The model accurately distinguished water body areas and showed lower accuracy in wetlands than in the other LULC types. The effect of the DEM on the accuracy of LULC was analyzed. When assessing the accuracy with respect to the incidence angle, owing to the radar shadow caused by the side-looking system of the SAR sensor, the OA tended to decrease as the incidence angle increased. This study is the first to use only KOMPSAT-5 single co-polarized data and deep learning methods to demonstrate the possibility of high-performance LULC monitoring. This study contributes to Earth surface monitoring and the development of deep learning approaches using the KOMPSAT-5 data.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.136-136
• /
• 2021

In this study, a deep convolutional neural network (DCNN) model is proposed for short-term precipitation forecasting using weather radar-based images. The DCNN model is a combination of convolutional neural networks, autoencoder neural networks, and U-net architecture. The weather radar-based image data used here are retrieved from competition for rainfall forecasting in Korea (AI Contest for Rainfall Prediction of Hydroelectric Dam Using Public Data), organized by Dacon under the sponsorship of the Korean Water Resources Association in October 2020. This data is collected from rainy events during the rainy season (April - October) from 2010 to 2017. These images have undergone a preprocessing step to convert from weather radar data to grayscale image data before they are exploited for the competition. Accordingly, each of these gray images covers a spatial dimension of 120×120 pixels and has a corresponding temporal resolution of 10 minutes. Here, each pixel corresponds to a grid of size 4km×4km. The DCNN model is designed in this study to provide 10-minute predictive images in advance. Then, precipitation information can be obtained from these forecast images through empirical conversion formulas. Model performance is assessed by comparing the Score index, which is defined based on the ratio of MAE (mean absolute error) to CSI (critical success index) values. The competition results have demonstrated the impressive performance of the DCNN model, where the Score value is 0.530 compared to the best value from the competition of 0.500, ranking 16^th out of 463 participating teams. This study's findings exhibit the potential of applying the DCNN model to short-term rainfall prediction using weather radar-based images. As a result, this model can be applied to other areas with different spatiotemporal resolutions.

• Journal of the Korean Society of Industry Convergence
• /
• v.23 no.3
• /
• pp.481-489
• /
• 2020

Safety and security during the handling of hazardous materials is a great concern for anyone in the field. One driving point in the security field is the ability to detect the source of the danger and take action against it as quickly as possible. Via the usage of a fully convolutional network, it is possible to create the label map of an input image, indicating what object is occupying the specific area of the image. This research employs the usage of U-net, which was constructed in biomedical field segmentation to segment cells, instead of the original FCN. One of the challenges that this research faces is the availability of ground truth with precise labeling for the dataset. Testing the network after training resulted in some images where the network pronounces even better detail than the expected label map. With better detailed label map, the network might be able to produce better segmentation is something to be studied in further research.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.49-52
• /
• 2022

건물 외벽에 발생하는 균열은 시설물 구조 안전에 영향을 미치며 그 크기에 따라 위험도가 달라진다. 이에 따라 전문검사관의 현장 점검을 통해 발생 균열 두께를 정밀하게 측정할 필요가 있고 최근에는 이러한 현장 안전점검에 인공지능을 도입하려는 추세다. 그러나 기존의 균열 데이터셋은 주로 콘크리트에만 한정되어 다양한 외벽에 강인한 모델을 구축하기 어렵고 균열 두께를 측정하기 위해 정확한 마스크(Mask) 정보가 필요하나 이를 만족하는 데이터셋이 부재하다. 본 논문에서는 다양한 외벽에 강인한 균열 구획화 모델을 목적으로 2,744장의 이미지를 촬영하고 매직 완드 기법으로 라벨링을 진행해 데이터셋을 구축 후, 이를 바탕으로 딥러닝 기반 균열 구획화 모델을 개발했다. UNet-ResNet50을 최종모델로 선정 및 개발 결과, 테스트 데이터셋에 대해 81.22%의 class IoU 성능을 보였다. 본 연구의 기술을 바탕으로 균열 두께를 측정하여 건축물 안전점검에 활용될 수 있기를 기대한다.