• Archives of Plastic Surgery
• /
• 제43권6호
• /
• pp.586-589
• /
• 2016

A pressure sore wound is often extensive or complicated by local infection involving adjacent soft tissue and bone. In this case, a regional flap after simple debridement is not adequate. Here, we present a case of an extensive pressure sore in the sacral area with deep tissue infection. A 43-year-old female patient with a complicated sore with deep tissue infection had a presacral abscess, an iliopsoas abscess, and an epidural abscess in the lumbar spine. After a multidisciplinary approach performed in stages, the infection had subsided and removal of the devitalized tissue was possible. The large soft tissue defect with significant depth was reconstructed with a free latissimus dorsi musculocutaneous flap, which was expected to act as a local barrier from vertical infection and provide tensionless skin coverage upon hip flexion. The extensive sacral sore was treated effectively without complication, and the deep tissue infection completely resolved. There was no evidence of donor site morbidity, and wheelchair ambulation was possible by a month after surgery.

• Journal of Advanced Research in Ocean Engineering
• /
• 제2권1호
• /
• pp.1-11
• /
• 2016

The objective of this paper is to present an analytical solution in deep water waves and verify the validity of the theory (Shin, 2015). Hence this is a follow-up to Shin (2015). Instead of a variational approach, another approach was considered for a more accurate assessment in this study. The products of two coefficients were not neglected in this study. The two wave profiles from the KFSBC and DFSBC were evaluated at N discrete points on the free-surface, and the combination coefficients were determined for when the two curves pass the discrete points. Thus, the solution satisfies the differential equation (DE), bottom boundary condition (BBC), and the kinematic free surface boundary condition (KFSBC) exactly. The error in the dynamic free surface boundary condition (DFSBC) is less than 0.003%. The wave theory was simplified based on the assumption tanh $D{\approx}1$ in this paper. Unlike the perturbation method, the results are possible for steep waves and can be calculated without iteration. The result is very simple compared to the 5th Stokes' theory. Stokes' breaking-wave criterion has been checked in this study.

• 대한원격탐사학회지
• /
• 제37권1호
• /
• pp.111-122
• /
• 2021

Building footprint extraction is an active topic in the domain of remote sensing, since buildings are a fundamental unit of urban areas. Deep convolutional neural networks successfully perform footprint extraction from optical satellite images. However, semantic segmentation produces coarse results in the output, such as blurred and rounded boundaries, which are caused by the use of convolutional layers with large receptive fields and pooling layers. The objective of this study is to generate visually enhanced building objects by directly extracting the vertices of individual buildings by combining instance segmentation and keypoint detection. The target keypoints in building extraction are defined as points of interest based on the local image gradient direction, that is, the vertices of a building polygon. The proposed framework follows a two-stage, top-down approach that is divided into object detection and keypoint estimation. Keypoints between instances are distinguished by merging the rough segmentation masks and the local features of regions of interest. A building polygon is created by grouping the predicted keypoints through a simple geometric method. Our model achieved an F1-score of 0.650 with an mIoU of 62.6 for building footprint extraction using the OpenCitesAI dataset. The results demonstrated that the proposed framework using keypoint estimation exhibited better segmentation performance when compared with Mask R-CNN in terms of both qualitative and quantitative results.

• 한국정보통신학회논문지
• /
• 제25권3호
• /
• pp.486-489
• /
• 2021

Recently, the demand for the use of personal mobility vehicles, such as an electric kickboard, is increasing explosively because of its high portability and usability. However, the number of traffic accidents caused by personal mobility vehicles has also increased rapidly in recent years. To address the issues regarding the driver's safety, we propose a novel approach that can monitor context information around personal mobility vehicles using deep learning-based object detection and smartphone captured videos. In the proposed framework, a smartphone is attached to a personal mobility device and a front or rear view is recorded to detect an approaching object that may affect the driver's safety. Through the detection results using YOLOv5 model, we report the preliminary results and validated the feasibility of the proposed approach.

• Imaging Science in Dentistry
• /
• 제52권3호
• /
• pp.275-281
• /
• 2022

Purpose: The aim of this study was to assess the performance of a deep learning system for permanent tooth germ detection on pediatric panoramic radiographs. Materials and Methods: In total, 4518 anonymized panoramic radiographs of children between 5 and 13 years of age were collected. YOLOv4, a convolutional neural network (CNN)-based object detection model, was used to automatically detect permanent tooth germs. Panoramic images of children processed in LabelImg were trained and tested in the YOLOv4 algorithm. True-positive, false-positive, and false-negative rates were calculated. A confusion matrix was used to evaluate the performance of the model. Results: The YOLOv4 model, which detected permanent tooth germs on pediatric panoramic radiographs, provided an average precision value of 94.16% and an F1 value of 0.90, indicating a high level of significance. The average YOLOv4 inference time was 90 ms. Conclusion: The detection of permanent tooth germs on pediatric panoramic X-rays using a deep learning-based approach may facilitate the early diagnosis of tooth deficiency or supernumerary teeth and help dental practitioners find more accurate treatment options while saving time and effort

• Journal of Information Processing Systems
• /
• 제19권3호
• /
• pp.355-369
• /
• 2023

Since the widespread adoption of deep-learning and related distributed representation, there have been substantial advancements in part-of-speech (POS) tagging for many languages. When training word representations, morphology and shape are typically ignored, as these representations rely primarily on collecting syntactic and semantic aspects of words. However, for tasks like POS tagging, notably in morphologically rich and resource-limited language environments, the intra-word information is essential. In this study, we introduce a deep neural network (DNN) for POS tagging that learns character-level word representations and combines them with general word representations. Using the proposed approach and omitting hand-crafted features, we achieve 90.47%, 80.16%, and 79.32% accuracy on our own dataset for three morphologically rich languages: Uyghur, Uzbek, and Kyrgyz. The experimental results reveal that the presented character-based strategy greatly improves POS tagging performance for several morphologically rich languages (MRL) where character information is significant. Furthermore, when compared to the previously reported state-of-the-art POS tagging results for Turkish on the METU Turkish Treebank dataset, the proposed approach improved on the prior work slightly. As a result, the experimental results indicate that character-based representations outperform word-level representations for MRL performance. Our technique is also robust towards the-out-of-vocabulary issues and performs better on manually edited text.

• 로봇학회논문지
• /
• 제19권1호
• /
• pp.117-129
• /
• 2024

This paper introduces collective navigation through a narrow gap using a curriculum-based deep reinforcement learning algorithm for a swarm of unmanned aerial vehicles (UAVs). Collective navigation in complex environments is essential for various applications such as search and rescue, environment monitoring and military tasks operations. Conventional methods, which are easily interpretable from an engineering perspective, divide the navigation tasks into mapping, planning, and control; however, they struggle with increased latency and unmodeled environmental factors. Recently, learning-based methods have addressed these problems by employing the end-to-end framework with neural networks. Nonetheless, most existing learning-based approaches face challenges in complex scenarios particularly for navigating through a narrow gap or when a leader or informed UAV is unavailable. Our approach uses the information of a certain number of nearest neighboring UAVs and incorporates a task-specific curriculum to reduce learning time and train a robust model. The effectiveness of the proposed algorithm is verified through an ablation study and quantitative metrics. Simulation results demonstrate that our approach outperforms existing methods.

• 한국콘크리트학회:학술대회논문집
• /
• 한국콘크리트학회 2003년도 봄 학술발표회 논문집
• /
• pp.918-923
• /
• 2003

Reinforced concrete deep beams are commonly used in many structural applications, including transfer girders, pile caps, foundation walls, and offshore structures. The existing design methods were developed and calibrated using normal strength concrete test results, and their applicability th HSC deep beams must be assessed. For the shear strength prediction of high-strength concrete(HSC) deep beams, this paper proposed Softened Strut-and-Tie Model(SSTM) considered HSC and bending moment effect. The shear strength predictions of the refined model, the formulas the ACI 318-02 Appendix A STM, and Eq. of ACI 318-99 11.8 are compared with the collected experimental data of 74 HSC deep beams with compressive strength in the range of 49-78MPa . It is shown the shear strength of deep beam calculated by those equations are conservative on comparing test results. The comparison shows that the performance of the proposed SSTM is better than the ACI Code approach for all the parameters under comparison. The parameters reviewed include concrete strength, the shear span-depth ratio, and the ratio of horizontal and vertical reinforcement. The proposed SSTM gave a mean predicted to experimental ratio of 0.99, 32 percent higher than ACI 318-02 Code, however with the low coefficient variation.

• Computers and Concrete
• /
• 제14권1호
• /
• pp.19-40
• /
• 2014

Reinforced concrete deep beams are structural beams with low shear span-to-depth ratio, and hence in which the strain distribution is significantly nonlinear and the conventional beam theory is not applicable. A strut-and-tie model is considered one of the most rational and simplest methods available for shear strength prediction and design of deep beams. The strut-and-tie model approach describes the shear failure of a deep beam using diagonal strut and truss mechanism: The diagonal strut mechanism represents compression stress fields that develop in the concrete web between diagonal cracks of the concrete while the truss mechanism accounts for the contributions of the horizontal and vertical web reinforcements. Based on a database of 406 experimental observations, this paper proposes a new strut-and-tie-model for accurate prediction of shear strength of reinforced concrete deep beams, and further improves the model by correcting the bias and quantifying the scatter using a Bayesian parameter estimation method. Seven existing deterministic models from design codes and the literature are compared with the proposed method. Finally, a limit-state design formula and the corresponding reduction factor are developed for the proposed strut-andtie model.

• 한국멀티미디어학회논문지
• /
• 제20권5호
• /
• pp.769-781
• /
• 2017

Deep learning networks like Convolutional Neural Networks (CNNs) show successful performances in many computer vision applications such as image classification, object detection, and so on. For implementation of deep learning networks in embedded system with limited processing power and memory, deep learning network may need to be simplified. However, simplified deep learning network cannot learn every possible scene. One realistic strategy for embedded deep learning network is to construct a simplified deep learning network model optimized for the scene images of the installation place. Then, automatic training will be necessitated for commercialization. In this paper, as an intermediate step toward automatic training under fisheye camera environments, we study more precise human localization in fisheye images, and propose an accurate human localization method, Automatic Ground-Truth Labelling Method (AGTLM). AGTLM first localizes candidate human object bounding boxes by utilizing GoogLeNet-LSTM approach, and after reassurance process by GoogLeNet-based CNN network, finally refines them more correctly and precisely(tightly) by applying saliency object detection technique. The performance improvement of the proposed human localization method, AGTLM with respect to accuracy and tightness is shown through several experiments.