• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.181-193
• /
• 2022

Data-driven structural health monitoring (SHM) of civil infrastructure can be used to continuously assess the state of a structure, allowing preemptive safety measures to be carried out. Long-term monitoring of large-scale civil infrastructure often involves data-collection using a network of numerous sensors of various types. Malfunctioning sensors in the network are common, which can disrupt the condition assessment and even lead to false-negative indications of damage. The overwhelming size of the data collected renders manual approaches to ensure data quality intractable. The task of detecting and classifying an anomaly in the raw data is non-trivial. We propose an approach to automate this task, improving upon the previously developed technique of image-based pre-processing on one-dimensional (1D) data by enriching the features of the neural network input data with multiple channels. In particular, feature engineering is employed to convert the measured time histories into a 3-channel image comprised of (i) the time history, (ii) the spectrogram, and (iii) the probability density function representation of the signal. To demonstrate this approach, a CNN model is designed and trained on a dataset consisting of acceleration records of sensors installed on a long-span bridge, with the goal of fault detection and classification. The effect of imbalance in anomaly patterns observed is studied to better account for unseen test cases. The proposed framework achieves high overall accuracy and recall even when tested on an unseen dataset that is much larger than the samples used for training, offering a viable solution for implementation on full-scale structures where limited labeled-training data is available.

• Physical Therapy Korea
• /
• v.22 no.3
• /
• pp.33-40
• /
• 2015

The purpose of this study was to investigate the effect of multi joint-joint position sense (MJ-JPS) training on joint position sense, balance, and gait ability in stroke patients. A total of 18 stroke patients participated in the study. The subjects were allocated randomly into two groups: an experimental group and a control group. Participants in the experimental group received MJ-JPS training (10 min) and conventional treatment (20 min), but participants in the control group only received conventional treatment (30 min). Both groups received training for five times per week for six weeks. MJ-JPS is a training method used to increase proprioception in the lower extremities; as such, it is used, to position the lower extremities in a given space. MJ-JPS measurement was captured via video using a Image J program to calculate the error distance. Balance ability was measured using Timed Up and Go (TUG) and the Berg Balance Scale (BBS). Gait ability was measured with a 10 m walking test (10MWT) and by climbing four flights of stairs. The Shapiro-Wilk test was used to assess normalization. Within-group differences were analyzed using the paired t-test. Between-group differences were analyzed using the independent t-test. The experimental group showed a significant decrease in error distance (MJ-JPS) compared to the control group (p<.05). Both groups showed a significant difference in their BBS and 10MWT results (p<.05). The experimental group showed a significant decrease in their TUG and climbing results (p<.05), but the control group results for those two tasks were not found to be significant (p>.05). There was significant difference in MJ-JPS and by climbing four flights of stairs on variation of pre and post test in between groups (p<.05), but TUG and BBS and 10MWT was no significantly (p>.05). We suggest that the MJ-JPS training proposed in this study be used as an intervention to help improve the functional activity of the lower extremities in stroke patients.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.3
• /
• pp.1434-1449
• /
• 2019

Video object segmentation is a significant task in computer vision, but its performance is not very satisfactory. A method of video object segmentation using weakly temporal information is presented in this paper. Motivated by the phenomenon in reality that the motion of the object is a continuous and smooth process and the appearance of the object does not change much between adjacent frames in the video sequences, we use a feed-forward architecture with motion estimation to predict the mask of the current frame. We extend an additional mask channel for the previous frame segmentation result. The mask of the previous frame is treated as the input of the expanded channel after processing, and then we extract the temporal feature of the object and fuse it with other feature maps to generate the final mask. In addition, we introduce multi-mask guidance to improve the stability of the model. Moreover, we enhance segmentation performance by further training with the masks already obtained. Experiments show that our method achieves competitive results on DAVIS-2016 on single object segmentation compared to some state-of-the-art algorithms.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.3
• /
• pp.670-684
• /
• 2024

3D Cross-Modal Retrieval (3DCMR) is a task that retrieves 3D objects regardless of modalities, such as images, meshes, and point clouds. One of the most prominent methods used for 3DCMR is the Cross-Modal Center Loss Function (CLF) which applies the conventional center loss strategy for 3D cross-modal search and retrieval. Since CLF is based on center loss, the center features in CLF are also susceptible to subtle changes in hyperparameters and external inferences. For instance, performance degradation is observed when the batch size is too small. Furthermore, the Mean Squared Error (MSE) used in CLF is unable to adapt to changes in batch size and is vulnerable to data variations that occur during actual inference due to the use of simple Euclidean distance between multi-modal features. To address the problems that arise from small batch training, we propose a Noisy Center Loss (NCL) method to estimate the optimal center features. In addition, we apply the simple Siamese representation learning method (SimSiam) during optimal center feature estimation to compare projected features, making the proposed method robust to changes in batch size and variations in data. As a result, the proposed approach demonstrates improved performance in ModelNet40 dataset compared to the conventional methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.1
• /
• pp.50-56
• /
• 2020

With the rapid development of deep neural network theory and application research, the effect of face detection has been improved. However, due to the complexity of deep neural network calculation and the high complexity of the detection environment, how to detect face quickly and accurately becomes the main problem. This paper is based on the relatively simple model of the MTCNN model, using FDDB (Face Detection Dataset and Benchmark Homepage), LFW (Field Label Face) and FaceScrub public datasets as training samples. At the same time of sorting out and introducing MTCNN(Multi-Task Cascaded Convolutional Neural Network) model, it explores how to improve training speed and Increase performance at the same time. In this paper, the dynamic image pyramid technology is used to replace the traditional image pyramid technology to segment samples, and OHEM (the online hard example mine) function in MTCNN model is deleted in training, so as to improve the training speed.

• Journal of the Institute of Convergence Signal Processing
• /
• v.3 no.4
• /
• pp.62-66
• /
• 2002

An attempt of using SOCMAC neural network for the prediction of a nonlinear sequence, which is generated by Mackey-Glass equation, is reported. The ,report shows the SOCMAC can handle a system with multi-dimensional continuous inputs, which has been considered very difficult, if not impossible, task to be implemented by a CMAC neural network because of a huge amount of memory required. Also, an improved training method based on the variable receptive fields is proposed. The Performance ranged somewhere around those of TDNN and BP neural networks.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2021.06a
• /
• pp.105-108
• /
• 2021

Generative adversarial networks (GANs) have reached a great result at creating the synthesis image, especially in the face generation task. Unlike other deep learning tasks, the input of GANs is usually the random vector sampled by a probability distribution, which leads to unstable training and unpredictable output. One way to solve those problems is to employ the label condition in both the generator and discriminator. CelebA and FFHQ are the two most famous datasets for face image generation. While CelebA contains attribute annotations for more than 200,000 images, FFHQ does not have attribute annotations. Thus, in this work, we introduce a method to learn the attributes from CelebA then predict both soft and hard labels for FFHQ. The evaluated result from our model achieves 0.7611 points of the metric is the area under the receiver operating characteristic curve.

• The korean journal of orthodontics
• /
• v.51 no.2
• /
• pp.77-85
• /
• 2021

Objective: To evaluate the accuracy of a multi-stage convolutional neural network (CNN) model-based automated identification system for posteroanterior (PA) cephalometric landmarks. Methods: The multi-stage CNN model was implemented with a personal computer. A total of 430 PA-cephalograms synthesized from cone-beam computed tomography scans (CBCT-PA) were selected as samples. Twenty-three landmarks used for Tweemac analysis were manually identified on all CBCT-PA images by a single examiner. Intra-examiner reproducibility was confirmed by repeating the identification on 85 randomly selected images, which were subsequently set as test data, with a two-week interval before training. For initial learning stage of the multi-stage CNN model, the data from 345 of 430 CBCT-PA images were used, after which the multi-stage CNN model was tested with previous 85 images. The first manual identification on these 85 images was set as a truth ground. The mean radial error (MRE) and successful detection rate (SDR) were calculated to evaluate the errors in manual identification and artificial intelligence (AI) prediction. Results: The AI showed an average MRE of 2.23 ± 2.02 mm with an SDR of 60.88% for errors of 2 mm or lower. However, in a comparison of the repetitive task, the AI predicted landmarks at the same position, while the MRE for the repeated manual identification was 1.31 ± 0.94 mm. Conclusions: Automated identification for CBCT-synthesized PA cephalometric landmarks did not sufficiently achieve the clinically favorable error range of less than 2 mm. However, AI landmark identification on PA cephalograms showed better consistency than manual identification.

• The Journal of Korean Physical Therapy
• /
• v.28 no.1
• /
• pp.8-13
• /
• 2016

Purpose: The purpose of the present current study was to examine control of upper limb multi-joint movements with differential coordination stability. To achieve the goals of the study, torque analyses were utilized to answer questions about how torque components were differed among various elbow-wrist coordination patterns. Methods: Eight self-reported right-handed college students (3 males and 5 females, mean age=20.6 yr) were volunteered. The task required participants to rhythmically coordinate the flexion-extension motions of their elbow and wrist with coordination relationship of $0^{\circ}$, $90^{\circ}$, and $180^{\circ}$relative phases between the two joints. Mean relative phase and phase stability (standard deviation of relative phase) were computed to for analysisze of overall coordination performance. To determine the figure out characteristics of torque components in elbow and wrist joints, impulse values of muscle torque (MT) and interactive torque (IT) and MT as a percentage of cycle duration (MT-PCD) were analyzed. Results: Torque results showed that the proximal elbow joint generated motions with mainly muscle efforts regardless of coordination patterns, while the distal wrist joint adjusted the coordination patterns by changing amount of MT. Impulse analyses showed that the least stable $90^{\circ}$ pattern was performed by utilizing a similar coordination strategy of the most stable $0^{\circ}$ pattern. Conclusion: The present current study suggests that the roles of distal and proximal joints differ in order to achieve various multi-joint coordination movements. This study provides information for use in gives an idea to development of rehabilitation or training programs for to persons with an impaired upper limb motor ability.

• The Korean Journal of Petroleum Geology
• /
• v.7 no.1_2 s.8
• /
• pp.19-27
• /
• 1999

Korea, an energy-resources-poor country, imports $100{\%}$ of its, oil and, natural gas supply, which accounts for the greater part of its total primary requirements. One of the important task of the government is diversification of available energy resources such as oil and natural gas. Natural gas hydrate, which is non-conventional types of natural gas, distributes worldwide, especially in marine and permafrost. It would become a target of natural gas resources in the near future. Especially sigrificant amount of hydrates are expected to be located in the East Sea around Korea Peninsular. This paper describes about a multi-year overall project framework of basic research and technological development of natural gas hydrate in Korea focused on the interpretation of the seismic survey, the characteristics and physical properties of the natural gas hydrate, and the utilizable technology of natural gas hydrates from the status of research and development of the world.