• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.25 no.6
• /
• pp.25-33
• /
• 2022

In this study, a monitoring method using high-resolution images acquired by unmanned aerial vehicles and deep learning algorithms was proposed for the management of the Sinduri coastal sand dunes. Class classification was done using U-net, a semantic division method. The classification target classified 3 types of sand dune vegetation into 4 classes, and the model was trained and tested with a total of 320 training images and 48 test images. Ignored label was applied to improve the performance of the model, and then evaluated by applying two loss functions, CE Loss and BCE Loss. As a result of the evaluation, when CE Loss was applied, the value of mIoU for each class was the highest, but it can be judged that the performance of BCE Loss is better considering the time efficiency consumed in learning. It is meaningful as a pilot application of unmanned aerial vehicles and deep learning as a method to monitor and manage sand dune vegetation. The possibility of using the deep learning image analysis technology to monitor sand dune vegetation has been confirmed, and it is expected that the proposed method can be used not only in sand dune vegetation but also in various fields such as forests and grasslands.

• 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.

• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.351-363
• /
• 2023

Visual structural inspections are an inseparable part of post-earthquake damage assessments. With unmanned aerial vehicles (UAVs) establishing a new frontier in visual inspections, there are major computational challenges in processing the collected massive amounts of high-resolution visual data. We propose twin deep learning models that can provide accurate high-resolution structural components and damage segmentation masks efficiently. The traditional approach to cope with high memory computational demands is to either uniformly downsample the raw images at the price of losing fine local details or cropping smaller parts of the images leading to a loss of global contextual information. Therefore, our twin models comprising Trainable Resizing for high-resolution Segmentation Network (TRS-Net) and DmgFormer approaches the global and local semantics from different perspectives. TRS-Net is a compound, high-resolution segmentation architecture equipped with learnable downsampler and upsampler modules to minimize information loss for optimal performance and efficiency. DmgFormer utilizes a transformer backbone and a convolutional decoder head with skip connections on a grid of crops aiming for high precision learning without downsizing. An augmented inference technique is used to boost performance further and reduce the possible loss of context due to grid cropping. Comprehensive experiments have been performed on the 3D physics-based graphics models (PBGMs) synthetic environments in the QuakeCity dataset. The proposed framework is evaluated using several metrics on three segmentation tasks: component type, component damage state, and global damage (crack, rebar, spalling). The models were developed as part of the 2^nd International Competition for Structural Health Monitoring.

• Smart Structures and Systems
• /
• v.31 no.4
• /
• 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.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.1
• /
• pp.463-469
• /
• 2023

The aim of this study is to provide basic data that can contribute to improving successful clinical adaptation and reducing turnover of new graduate nurses by analyzing research related to reality shock experienced by new graduate nurses using text network analysis. The topics of reality shock experienced by new graduate nurses were extracted from 115 papers published in domestic and foreign journals from January 2002 to December 2021. Articles were retrieved from 6 databases (Korean DB: DBpia, KISS, RISS /International DB: Web of science, Springer, Scopus). Keywords were extracted from the abstract and organized using semantic morphemes. Network analysis and topic modeling for subject knowledge structure analysis were performed using NetMiner 4.5.0 program. The core keywords included 'new graduate nurses', 'reality shock', 'transition', 'student nurse', 'experience', 'practice', 'work environment', 'role', 'care' and 'education'. In recent articles on reality shock experienced by new graduate nurses, three major topics were extracted by LDA (Latent Dirichlet Allocation) techniques: 'turnover', 'work environment', 'experience of transition'. Based on this research, the necessity of interventional research that can effectively reduce the reality shock experienced by new graduate nurses and successfully help clinical adaptation is suggested.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.26 no.6
• /
• pp.85-96
• /
• 2023

This study proposes a method for forest vegetation monitoring using high-resolution aerial imagery captured by unmanned aerial vehicles(UAV) and deep learning technology. The research site was selected in the forested area of Mountain Dogo, Asan City, Chungcheongnam-do, and the target species for monitoring included Pinus densiflora, Quercus mongolica, and Quercus acutissima. To classify vegetation species at the pixel level in UAV imagery based on characteristics such as leaf shape, size, and color, the study employed the semantic segmentation method using the prominent U-net deep learning model. The research results indicated that it was possible to visually distinguish Pinus densiflora Siebold & Zucc, Quercus mongolica Fisch. ex Ledeb, and Quercus acutissima Carruth in 135 aerial images captured by UAV. Out of these, 104 images were used as training data for the deep learning model, while 31 images were used for inference. The optimization of the deep learning model resulted in an overall average pixel accuracy of 92.60, with mIoU at 0.80 and FIoU at 0.82, demonstrating the successful construction of a reliable deep learning model. This study is significant as a pilot case for the application of UAV and deep learning to monitor and manage representative species among climate-vulnerable vegetation, including Pinus densiflora, Quercus mongolica, and Quercus acutissima. It is expected that in the future, UAV and deep learning models can be applied to a variety of vegetation species to better address forest management.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2001.01a
• /
• pp.285-291
• /
• 2001

This paper describes a natural language question answering system that can be used by students in getting as solution to their queries. Unlike AI question answering system that focus on the generation of new answers, the present system retrieves existing ones from question-answer files. Unlike information retrieval approaches that rely on a purely lexical metric of similarity between query and document, it uses a semantic knowledge base (WordNet) to improve its ability to match question. Paper describes the design and the current implementation of the system as an intelligent tutoring system. Main drawback of the existing tutoring systems is that the computer poses a question to the students and guides them in reaching the solution to the problem. In the present approach, a student asks any question related to the topic and gets a suitable reply. Based on his query, he can either get a direct answer to his question or a set of questions (to a maximum of 3 or 4) which bear the greatest resemblance to the user input. We further analyze-application fields for such kind of a system and discuss the scope for future research in this area.

• Journal of KIISE:Software and Applications
• /
• v.27 no.3
• /
• pp.302-313
• /
• 2000

In English-Korean machine translation, transitive verb-objective collocation is utilized for accurate translation of an English verbal phrase into Korean. This paper presents an algorithm for correct verb translation based on the k-nearest neighbor learning. The semantic distance is defined on the WordNet for the k-nearest neighbor learning. And we also present algorithms for automatic collocation dictionary optimization. The algorithms extract transitive verb-objective pairs as training examples from large corpora and minimize the examples, considering the tradeoff between translation accuracy and example size. Experiments show that these algorithms optimized collocation dictionary keeping about 90% accuracy for a verb 'build'.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.06c
• /
• pp.189-193
• /
• 2007

개념 간의 유사도 측정 방법은 의미망에서의 두 개념의 최단 경로의 수 노드의 깊이 관계의 종류 등의 정보를 이용하는 링크(Link) 기반 방법, 대용량의 말뭉치에서의 개념의 발생빈도를 확률로 계산한 정보량(Information Content) 기반 방법, 관련 단어들의 공기정보를 활용한 의미(Gloss) 기반 방법이 있으며, 이미 국외에서는 WordNet과 같은 의미적 언어자원을 활용하여 많은 연구가 진행되고 있다. 그러나 국내에서는 아직 한국어 의미망을 바탕으로 한 개념간의 유사성 측정 방법이나 이를 활용하는 방법에 대한 연구가 미흡하다. 본 논문에서는 이를 바탕으로 링크 타입 노드의 깊이 최단경로 정보량 등의 요소를 이용한 의미 유사도 측정방법을 제안하고 이를 활용하여 명사-용언간의 연계 정보를 확보함으로써, 효율적으로 명사-용언간의 네트워크를 구축하도록 한다.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2006.06a
• /
• pp.229-233
• /
• 2006

기계번역(Machine Translaton, MT), 다국어 정보 검색, 의미 정보 검색 등에 대한 연구는 시소러스, 지식베이스, 사전 검색, 의미망, 코퍼스등과 같은 다양한 방법으로 이루어지고 있다. 시맨틱 웹이 등장과 시맨틱 웹 기반 기술의 발전에 따라 위 연구들을 시맨틱 웹에 적용시킬 필요성도 제안되었다. 특히 한국어 시소러스, 워드넷(WordNet), 전자 세종 사전, 가도까와(Kadokawa) 시소러스와 같은 지식베이스가 개발되었으나 활용 분야에 따라 그 구축 방법론이 다르게 적용되어, 위 연구에 효과적으로 통용될 수 있는 지식베이스는 실질적으로 구축되지 못한 실정이다. 따라서 본 연구에서는 세종 사전과 가도까와 시소러스, 한/일 기계 번역 사전 그리고 전문 용어 사전을 기반으로 한국어와 일본어 지식베이스를 위한 사전 온톨로지 서버를 정의하여 의미 정보를 구성하고, Semantic Web Rule Markup Language (이하 SWRL)을 이용해 구문 정보 규칙을 정의한다. 그리고 SWRL 기반 정방향 추론 엔진을 이용하여 번역에 필요한 추론 엔진을 구성하고 문장 구문형성 규칙 추론 엔진을 통해 사용자에게 한국어와 일본어의 문장 구성 변환을 제공한다. 본 연구는 현재 기계 번역이 갖고 있는 다의성, 술부 어순의 차이, 경어체 등 아직 해결해야 할 많은 부분들에 대한 해결 방안으로서 시맨틱 웹 기반 기술과의 활용방안을 제시하고자 한다.