• Journal of the Korea Society of Computer and Information
• /
• v.29 no.5
• /
• pp.21-29
• /
• 2024

Reducing the radiation dose during CT scanning can lower the risk of radiation exposure, but not only does the image resolution significantly deteriorate, but the effectiveness of diagnosis is reduced due to the generation of noise. Therefore, noise removal from CT images is a very important and essential processing process in the image restoration. Until now, there are limitations in removing only the noise by separating the noise and the original signal in the image area. In this paper, we aim to effectively remove noise from CT images using the wavelet transform-based GAN model, that is, the WT-GAN model in the frequency domain. The GAN model used here generates images with noise removed through a U-Net structured generator and a PatchGAN structured discriminator. To evaluate the performance of the WT-GAN model proposed in this paper, experiments were conducted on CT images damaged by various noises, namely Gaussian noise, Poisson noise, and speckle noise. As a result of the performance experiment, the WT-GAN model is better than the traditional filter, that is, the BM3D filter, as well as the existing deep learning models, such as DnCNN, CDAE model, and U-Net GAN model, in qualitative and quantitative measures, that is, PSNR (Peak Signal-to-Noise Ratio) and SSIM (Structural Similarity Index Measure) showed excellent results.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.24 no.6
• /
• pp.89-96
• /
• 2021

In this study, a deep-learning image analysis model was established and validated for AI-based monitoring of the tidal flat ecosystem for marine protected creatures Ocypode stimpsoni and their habitat. The data in the study was constructed using an unmanned aerial vehicle, and the U-net model was applied for the deep learning model. The accuracy of deep learning model learning results was about 0.76 and about 0.8 each for the Ocypode stimpsoni and their burrow whose accuracy was higher. Analyzing the distribution of crabs and burrows by putting orthomosaic images of the entire study area to the learned deep learning model, it was confirmed that 1,943 Ocypode stimpsoni and 2,807 burrow were distributed in the study area. Through this study, the possibility of using the deep learning image analysis technology for monitoring the tidal ecosystem was confirmed. And it is expected that it can be used in the tidal ecosystem monitoring field by expanding the monitoring sites and target species in the future.

• Journal of Korea Water Resources Association
• /
• v.53 no.12
• /
• pp.1159-1172
• /
• 2020

This study applied deep convolution neural network based on U-Net and SegNet using long period weather radar data to very short-term rainfall prediction. And the results were compared and evaluated with the translation model. For training and validation of deep neural network, Mt. Gwanak and Mt. Gwangdeoksan radar data were collected from 2010 to 2016 and converted to a gray-scale image file in an HDF5 format with a 1km spatial resolution. The deep neural network model was trained to predict precipitation after 10 minutes by using the four consecutive radar image data, and the recursive method of repeating forecasts was applied to carry out lead time 60 minutes with the pretrained deep neural network model. To evaluate the performance of deep neural network prediction model, 24 rain cases in 2017 were forecast for rainfall up to 60 minutes in advance. As a result of evaluating the predicted performance by calculating the mean absolute error (MAE) and critical success index (CSI) at the threshold of 0.1, 1, and 5 mm/hr, the deep neural network model showed better performance in the case of rainfall threshold of 0.1, 1 mm/hr in terms of MAE, and showed better performance than the translation model for lead time 50 minutes in terms of CSI. In particular, although the deep neural network prediction model performed generally better than the translation model for weak rainfall of 5 mm/hr or less, the deep neural network prediction model had limitations in predicting distinct precipitation characteristics of high intensity as a result of the evaluation of threshold of 5 mm/hr. The longer lead time, the spatial smoothness increase with lead time thereby reducing the accuracy of rainfall prediction The translation model turned out to be superior in predicting the exceedance of higher intensity thresholds (> 5 mm/hr) because it preserves distinct precipitation characteristics, but the rainfall position tends to shift incorrectly. This study are expected to be helpful for the improvement of radar rainfall prediction model using deep neural networks in the future. In addition, the massive weather radar data established in this study will be provided through open repositories for future use in subsequent studies.

• Journal of IKEEE
• /
• v.26 no.2
• /
• pp.160-168
• /
• 2022

In this paper, we propose a 3D mesh reconstruction method from a single image using deep learning and a sphere shape transformation method. The proposed method has the following originality that is different from the existing method. First, the position of the vertex of the sphere is modified to be very similar to the 3D point cloud of an object through a deep learning network, unlike the existing method of building edges or faces by connecting nearby points. Because 3D point cloud is used, less memory is required and faster operation is possible because only addition operation is performed between offset value at the vertices of the sphere. Second, the 3D mesh is reconstructed by covering the surface information of the sphere on the modified vertices. Even when the distance between the points of the 3D point cloud created by correcting the position of the vertices of the sphere is not constant, it already has the face information of the sphere called face information of the sphere, which indicates whether the points are connected or not, thereby preventing simplification or loss of expression. can do. In order to evaluate the objective reliability of the proposed method, the experiment was conducted in the same way as in the comparative papers using the ShapeNet dataset, which is an open standard dataset. As a result, the IoU value of the method proposed in this paper was 0.581, and the chamfer distance value was It was calculated as 0.212. The higher the IoU value and the lower the chamfer distance value, the better the results. Therefore, the efficiency of the 3D mesh reconstruction was demonstrated compared to the methods published in other papers.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.68 no.2
• /
• pp.359-363
• /
• 2019

As commercial drones have been widely used, concerns for collision accidents with people and invading secured properties are emerging. The detection of drone is a challenging problem. The deep learning based object detection techniques for detecting drones have been applied, but limited to the specific cases such as detection of drones from bird and/or background. We have tried not only detection of drones, but classification of different drones with an end-to-end model. YOLOv2 is used as an object detection model. In order to supplement insufficient data by shooting drones, data augmentation from collected images is executed. Also transfer learning from ImageNet for YOLOv2 darknet framework is performed. The experimental results for drone detection with average IoU and recall are compared and analysed.

• Tunnel and Underground Space
• /
• v.33 no.6
• /
• pp.508-518
• /
• 2023

This paper presents a new approach for the automatic mapping of discontinuities in a tunnel face based on its 3D digital model reconstructed by LiDAR scan or photogrammetry techniques. The main idea revolves around the identification of discontinuity areas in the 3D digital model of a tunnel face by segmenting its 2D projected images using a deep-learning semantic segmentation model called U-Net. The proposed deep learning model integrates various features including the projected RGB image, depth map image, and local surface properties-based images i.e., normal vector and curvature images to effectively segment areas of discontinuity in the images. Subsequently, the segmentation results are projected back onto the 3D model using depth maps and projection matrices to obtain an accurate representation of the location and extent of discontinuities within the 3D space. The performance of the segmentation model is evaluated by comparing the segmented results with their corresponding ground truths, which demonstrates the high accuracy of segmentation results with the intersection-over-union metric of approximately 0.8. Despite still being limited in training data, this method exhibits promising potential to address the limitations of conventional approaches, which only rely on normal vectors and unsupervised machine learning algorithms for grouping points in the 3D model into distinct sets of discontinuities.

• Korean Journal of Remote Sensing
• /
• v.38 no.1
• /
• pp.103-110
• /
• 2022

Since satellite images generally include clouds in the atmosphere, it is essential to detect or mask clouds before satellite image processing. Clouds were detected using physical characteristics of clouds in previous research. Cloud detection methods using deep learning techniques such as CNN or the modified U-Net in image segmentation field have been studied recently. Since image segmentation is the process of assigning a label to every pixel in an image, precise pixel-based dataset is required for cloud detection. Obtaining accurate training datasets is more important than a network configuration in image segmentation for cloud detection. Existing deep learning techniques used different training datasets. And test datasets were extracted from intra-dataset which were acquired by same sensor and procedure as training dataset. Different datasets make it difficult to determine which network shows a better overall performance. To verify the effectiveness of the cloud detection network such as Cloud-Net, two types of networks were trained using the cloud dataset from KOMPSAT-3 images provided by the AIHUB site and the L8-Cloud dataset from Landsat8 images which was publicly opened by a Cloud-Net author. Test data from intra-dataset of KOMPSAT-3 cloud dataset were used for validating the network. The simulation results show that the network trained with KOMPSAT-3 cloud dataset shows good performance on the network trained with L8-Cloud dataset. Because Landsat8 and KOMPSAT-3 satellite images have different GSDs, making it difficult to achieve good results from cross-sensor validation. The network could be superior for intra-dataset, but it could be inferior for cross-sensor data. It is necessary to study techniques that show good results in cross-senor validation dataset in the future.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.2
• /
• pp.33-42
• /
• 2022

In this paper, we proposed a deep learning model based on multi-scale dilated convolution feature fusion for the segmentation of black ice region in road image to send black ice warning to drivers in real time. In the proposed multi-scale dilated convolution feature fusion network, different dilated ratio convolutions are connected in parallel in the encoder blocks, and different dilated ratios are used in different resolution feature maps, and multi-layer feature information are fused together. The multi-scale dilated convolution feature fusion improves the performance by diversifying and expending the receptive field of the network and by preserving detailed space information and enhancing the effectiveness of diated convolutions. The performance of the proposed network model was gradually improved with the increase of the number of dilated convolution branch. The mIoU value of the proposed method is 96.46%, which was higher than the existing networks such as U-Net, FCN, PSPNet, ENet, LinkNet. The parameter was 1,858K, which was 6 times smaller than the existing LinkNet model. From the experimental results of Jetson Nano, the FPS of the proposed method was 3.63, which can realize segmentation of black ice field in real time.

• Journal of Biomedical Engineering Research
• /
• v.41 no.2
• /
• pp.94-100
• /
• 2020

The purpose of this study was to present the models for classifying the wrist X-ray images by types and for segmenting the radius automatically in each image using deep learning and to verify the learned models. The data were a total of 904 wrist X-rays with the distal radius fracture, consisting of 472 anteroposterior (AP) and 432 lateral images. The learning model was the ResNet50 model for AP/lateral image classification, and the U-Net model for segmentation of the radius. In the model for AP/lateral image classification, 100.0% was showed in precision, recall, and F1 score and area under curve (AUC) was 1.0. The model for segmentation of the radius showed an accuracy of 99.46%, a sensitivity of 89.68%, a specificity of 99.72%, and a Dice similarity coefficient of 90.05% in AP images and an accuracy of 99.37%, a sensitivity of 88.65%, a specificity of 99.69%, and a Dice similarity coefficient of 86.05% in lateral images. The model for AP/lateral classification and the segmentation model of the radius learned through deep learning showed favorable performances to expect clinical application.

• Journal of the Society of Naval Architects of Korea
• /
• v.58 no.5
• /
• pp.303-313
• /
• 2021

Recently, quality control of the Liquefied Natural Gas Carrier (LNGC) cargo hold and block-erection interference areas using 3D scanners have been performed, focusing on large shipyards and the international association of classification societies. In this study, as a part of the research on LNGC cargo hold quality management advancement, a study on deep-learning-based scaffolding system 3D point cloud object detection and post-processing were conducted using a LNGC cargo hold 3D point cloud. The scaffolding system point cloud object detection is based on the PointNet deep learning architecture that detects objects using point clouds, achieving 70% prediction accuracy. In addition, the possibility of improving the accuracy of object detection through parameter adjustment is confirmed, and the standard of Intersection over Union (IoU), an index for determining whether the object is the same, is achieved. To avoid the manual post-processing work, the object detection architecture allows automatic task performance and can achieve stable prediction accuracy through supplementation and improvement of learning data. In the future, an improved study will be conducted on not only the flat surface of the LNGC cargo hold but also complex systems such as curved surfaces, and the results are expected to be applicable in process progress automation rate monitoring and ship quality control.