• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2019년도 춘계학술대회
• /
• pp.432-435
• /
• 2019

Convolutional Neural Networks (CNNs) have recently been gaining popularity in the medical image analysis field because of their image segmentation capabilities. In this paper, we present a CNN that performs automated brain tumor segmentations of sparsely annotated 3D Magnetic Resonance Imaging (MRI) scans. Our CNN is based on 3D U-net architecture, and it includes separate Dilated and Depth-wise Convolutions. It is fully-trained on the BraTS 2018 data set, and it produces more accurate results even when compared to the winners of the BraTS 2017 competition despite having a significantly smaller amount of parameters.

• Smart Structures and Systems
• /
• 제31권4호
• /
• 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.

• 대한의용생체공학회:의공학회지
• /
• 제41권1호
• /
• pp.48-54
• /
• 2020

In this paper, we deal with a liver fibrosis classification problem using ultrasound B-mode images. Commonly representative methods for classifying the stages of liver fibrosis include liver biopsy and diagnosis based on ultrasound images. The overall liver shape and the smoothness and roughness of speckle pattern represented in ultrasound images are used for determining the fibrosis stages. Although the ultrasound image based classification is used frequently as an alternative or complementary method of the invasive biopsy, it also has the limitations that liver fibrosis stage decision depends on the image quality and the doctor's experience. With the rapid development of deep learning algorithms, several studies using deep learning methods have been carried out for automated liver fibrosis classification and showed superior performance of high accuracy. The performance of those deep learning methods depends closely on the amount of datasets. We propose an enhanced U-net architecture to maximize the classification accuracy with limited small amount of image datasets. U-net is well known as a neural network for fast and precise segmentation of medical images. We design it newly for the purpose of classifying liver fibrosis stages. In order to assess the performance of the proposed architecture, numerical experiments are conducted on a total of 118 ultrasound B-mode images acquired from 78 patients with liver fibrosis symptoms of F0~F4 stages. The experimental results support that the performance of the proposed architecture is much better compared to the transfer learning using the pre-trained model of VGGNet.

• 한국공간정보시스템학회:학술대회논문집
• /
• 한국공간정보시스템학회 2005년도 GIS/RS 공동 춘계학술대회
• /
• pp.97-100
• /
• 2005

3차원 공간정보는 u-Korea, 전자정부, 유비쿼터스, LBS등의 기반 인프라 및 3차원 그래픽 처리기술, 가상현실 기술 등의 종합적으로 적용되는 고부가가치 통합 기술로 필요성이 부각되고 있다. 본 연구에서는 컴퓨터 그래픽스 분야에서 많이 적용되고 있는 공개 그래픽 라이브러리인 OpenGL(Open Graphics Library) 기반의 3D-GIS 가시화 어플리케이션 아키텍쳐를 중심으로 렌더링 기법을 분석하고자 한다. 한편 본 연구의 실험은 Visual Studio.NET환경에서 3D-GIS 모델 Prototype을 구현하여 수행하였으며 향후 실시간 모바일 3D-GIS 렌더링을 위한 기반 기술로 적용될 수 있는 OpcnGL-ES의 확장 가능성을 검토하고자 하였다.

• 한국멀티미디어학회논문지
• /
• 제22권11호
• /
• pp.1269-1279
• /
• 2019

In medical image analysis, segmentation is considered as a vital process since it partitions an image into coherent parts and extracts interesting objects from the image. In this paper, we consider automatic segmentations of OCT retinal images to find six layer boundaries using convolutional neural networks. Segmenting retinal images by layer boundaries is very important in diagnosing and predicting progress of eye diseases including diabetic retinopathy, glaucoma, and AMD (age-related macular degeneration). We applied well-known CNN architecture for general image segmentation, called Segnet, U-net, and CNN-S into this problem. We also proposed a shortest path-based algorithm for finding the layer boundaries from the outputs of Segnet and U-net. We analysed their performance on public OCT image data set. The experimental results show that the Segnet combined with the proposed shortest path-based boundary finding algorithm outperforms other two networks.

• 대한원격탐사학회지
• /
• 제39권6_3호
• /
• pp.1771-1777
• /
• 2023

최근 딥러닝의 발전으로 의미론적 분할을 통한 토지피복 분류 방법들이 제안되고 있다. 그러나 기존의 딥러닝 기반 모델들은 영상 정보만을 이용하기 때문에 시공간적 일관성을 담보할 수 없는 한계점이 있다. 이에 본 연구에서는 좌표 정보를 활용한 토지피복 분류 모델을 제안한다. 먼저 암시적 신경 표현 기법인 다중해상도 해시 인코더를 위경도 좌표계로 확장한 좌표 해시 인코더를 통해 좌표의 특징을 추출하였다. 다음으로 추출된 좌표 특징을 다양한 단계의 U-net 디코더와 결합하는 아키텍처를 제안하였다. 실험 결과, 제안 방법이 약 32% 향상된 분류 정확도를 보였고, 시공간적 일관성이 향상됨을 확인하였다.

• 한국정보기술학회논문지
• /
• 제17권9호
• /
• pp.99-112
• /
• 2019

도로 포장면의 크랙(crack)은 도로포장 구조의 열화를 입증하는 중요한 신호와 증상이다. 카메라 영상기반 도로포장 크랙 탐지는 강도 비균질성, 위상 복잡성, 낮은 대조도 및 노이즈성의 텍스처 배경 때문에 어려운 문제이다. 본 논문은 흑백영상에 대하여 깊은 신경망(DNN)에 기반하여 픽셀수준의 도로 크랙 탐지 및 분할 문제에 대해 다룬다. 변형된 U-net 네트워크와 고수준 특징 네트워크를 포함하는 새로운 DNN 구조를 제안한다. 본 연구의 중요 기여는 융합 층을 통해 공급되는 이들 네트워크의 결합 방법이다. 우리가 아는 한, 본 연구는 보도블럭 크랙 분할 및 탐지 문제를 결합을 소개한 최초의 논문이다. 크랙 탐지 및 분할의 시스템 성능은 새로운 구조를 사용하여 급격히 향상되었다. 제안된 시스템을 2개의 공개 데이터셋­크랙 포레스트 데이터셋(CFD)와 AigleRN 데이터셋­에 대하여 구현하고 평가하였다. 본 논문의 시스템은 여덟 가지의 최신 알고리즘과 같은 데이터셋으로 실험을 하였을 때, 가장 뛰어난 결과를 보여주었다.

• 대한조선학회논문집
• /
• 제58권5호
• /
• 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.

• Smart Structures and Systems
• /
• 제31권4호
• /
• pp.365-381
• /
• 2023

The existing vision-based techniques for inspection and condition assessment of civil infrastructure are mostly manual and consequently time-consuming, expensive, subjective, and risky. As a viable alternative, researchers in the past resorted to deep learning-based autonomous damage detection algorithms for expedited post-disaster reconnaissance of structures. Although a number of automatic damage detection algorithms have been proposed, the scarcity of labeled training data remains a major concern. To address this issue, this study proposed a semi-supervised learning (SSL) framework based on consistency regularization and cross-supervision. Image data from post-earthquake reconnaissance, that contains cracks, spalling, and exposed rebars are used to evaluate the proposed solution. Experiments are carried out under different data partition protocols, and it is shown that the proposed SSL method can make use of unlabeled images to enhance the segmentation performance when limited amount of ground truth labels are provided. This study also proposes DeepLab-AASPP and modified versions of U-Net++ based on channel-wise attention mechanism to better segment the components and damage areas from images of reinforced concrete buildings. The channel-wise attention mechanism can effectively improve the performance of the network by dynamically scaling the feature maps so that the networks can focus on more informative feature maps in the concatenation layer. The proposed DeepLab-AASPP achieves the best performance on component segmentation and damage state segmentation tasks with mIoU scores of 0.9850 and 0.7032, respectively. For crack, spalling, and rebar segmentation tasks, modified U-Net++ obtains the best performance with Igou scores (excluding the background pixels) of 0.5449, 0.9375, and 0.5018, respectively. The proposed architectures win the second place in IC-SHM2021 competition in all five tasks of Project 2.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.195-206
• /
• 2022

Crack detection is essential for inspection of existing structures and crack segmentation based on deep learning is a significant solution. However, datasets are usually one of the key issues. When building a new dataset for deep learning, laborious and time-consuming annotation of a large number of crack images is an obstacle. The aim of this study is to develop an approach that can automatically select a small portion of the most informative crack images from a large pool in order to annotate them, not to label all crack images. An active learning method with difficulty learning mechanism for crack segmentation tasks is proposed. Experiments are carried out on a crack image dataset of a steel box girder, which contains 500 images of 320×320 size for training, 100 for validation, and 190 for testing. In active learning experiments, the 500 images for training are acted as unlabeled image. The acquisition function in our method is compared with traditional acquisition functions, i.e., Query-By-Committee (QBC), Entropy, and Core-set. Further, comparisons are made on four common segmentation networks: U-Net, DeepLabV3, Feature Pyramid Network (FPN), and PSPNet. The results show that when training occurs with 200 (40%) of the most informative crack images that are selected by our method, the four segmentation networks can achieve 92%-95% of the obtained performance when training takes place with 500 (100%) crack images. The acquisition function in our method shows more accurate measurements of informativeness for unlabeled crack images compared to the four traditional acquisition functions at most active learning stages. Our method can select the most informative images for annotation from many unlabeled crack images automatically and accurately. Additionally, the dataset built after selecting 40% of all crack images can support crack segmentation networks that perform more than 92% when all the images are used.