• Journal of the Ergonomics Society of Korea
• /
• v.36 no.5
• /
• pp.385-394
• /
• 2017

Objective: The aim of this study is to investigate how to design effective virtual reality-based training (i.e., virtual training) in maritime safety and to present methods for enhancing interface fidelity by employing immersive interaction and 3D user interface (UI) design. Background: Emerging virtual reality technologies and hardware enable to provide immersive experiences to individuals. There is also a theory that the improvement of fidelity can improve the training efficiency. Such a sense of immersion can be utilized as an element for realizing effective training in the virtual space. Method: As an immersive interaction, we implemented gesture-based interaction using leap motion and Myo armband type sensors. Hand gestures captured from both sensors are used to interact with the virtual appliance in the scenario. The proposed 3D UI design is employed to visualize appropriate information for tasks in training. Results: A usability study to evaluate the effectiveness of the proposed method has been carried out. As a result, the usability test of satisfaction, intuitiveness of UI, ease of procedure learning, and equipment understanding showed that virtual training-based exercise was superior to existing training. These improvements were also independent of the type of input devices for virtual training. Conclusion: We have shown through experiments that the proposed interaction design results are more efficient interactions than the existing training method. The improvement of interface fidelity through intuitive and immediate feedback on the input device and the training information improve user satisfaction with the system, as well as training efficiency. Application: Design methods for an effective virtual training system can be applied to other areas by which trainees are required to do sophisticated job with their hands.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.49 no.4
• /
• pp.42-53
• /
• 2012

In pattern recognition and machine learning society, classification has been a classical problem and the most widely researched area. Adaptive boosting also known as Adaboost has been successfully applied to binary classification problems. It is a kind of boosting algorithm capable of constructing a strong classifier through a weighted combination of weak classifiers. On the other hand, the PCA and LDA algorithms are the most popular linear feature extraction methods used mainly for dimensionality reduction. In this paper, the combination of Adaboost and feature extraction methods is proposed for efficient classification of two class data. Conventionally, in classification problems, the roles of feature extraction and classification have been distinct, i.e., a feature extraction method and a classifier are applied sequentially to classify input variable into several categories. In this paper, these two steps are combined into one resulting in a good classification performance. More specifically, each projection vector is treated as a weak classifier in Adaboost algorithm to constitute a strong classifier for binary classification problems. The proposed algorithm is applied to UCI dataset and FRGC dataset and showed better recognition rates than sequential application of feature extraction and classification methods.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39C no.9
• /
• pp.811-819
• /
• 2014

Activity recognition is a key component in identifying the context of a user for providing services based on the application such as medical, entertainment and tactical scenarios. Instead of applying numerous sensor devices, as observed in many previous investigations, we are proposing the use of smartphone with its built-in multimodal sensors as an unobtrusive sensor device for recognition of six physical daily activities. As an improvement to previous works, accelerometer, gyroscope and magnetometer data are fused to recognize activities more reliably. The evaluation indicates that the IBK classifier using window size of 2s with 50% overlapping yields the highest accuracy (i.e., up to 99.33%). To achieve this peak accuracy, simple time-domain and frequency-domain features were extracted from raw sensor data of the smartphone.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_3
• /
• pp.979-995
• /
• 2023

As wildfires are difficult to predict, real-time monitoring is crucial for a timely response. Geostationary satellite images are very useful for active fire detection because they can monitor a vast area with high temporal resolution (e.g., 2 min). Existing satellite-based active fire detection algorithms detect thermal outliers using threshold values based on the statistical analysis of brightness temperature. However, the difficulty in establishing suitable thresholds for such threshold-based methods hinders their ability to detect fires with low intensity and achieve generalized performance. In light of these challenges, machine learning has emerged as a potential-solution. Until now, relatively simple techniques such as random forest, Vanilla convolutional neural network (CNN), and U-net have been applied for active fire detection. Therefore, this study proposed an active fire detection algorithm using state-of-the-art (SOTA) deep learning techniques using data from the Advanced Himawari Imager and evaluated it over East Asia and Australia. The SOTA model was developed by applying EfficientNet and lion optimizer, and the results were compared with the model using the Vanilla CNN structure. EfficientNet outperformed CNN with F1-scores of 0.88 and 0.83 in East Asia and Australia, respectively. The performance was better after using weighted loss, equal sampling, and image augmentation techniques to fix data imbalance issues compared to before the techniques were used, resulting in F1-scores of 0.92 in East Asia and 0.84 in Australia. It is anticipated that timely responses facilitated by the SOTA deep learning-based approach for active fire detection will effectively mitigate the damage caused by wildfires.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.1
• /
• pp.169-177
• /
• 2020

In South Korea, the results of R&D in science and technology are submitted to the National Science and Technology Information Service (NTIS) in reports that have Korea national science and technology standard classification codes (K-NSCC). However, considering there are more than 2000 sub-categories, it is non-trivial to choose correct classification codes without a clear understanding of the K-NSCC. In addition, there are few cases of automatic document classification research based on the K-NSCC, and there are no training data in the public domain. To the best of our knowledge, this study is the first attempt to build a highly performing K-NSCC classification system based on NTIS report meta-information from the last five years (2013-2017). To this end, about 210 mid-level categories were selected, and we conducted preprocessing considering the characteristics of research report metadata. More specifically, we propose a convolutional neural network (CNN) technique using only task names and keywords, which are the most influential fields. The proposed model is compared with several machine learning methods (e.g., the linear support vector classifier, CNN, gated recurrent unit, etc.) that show good performance in text classification, and that have a performance advantage of 1% to 7% based on a top-three F1 score.

• Journal of Korea Water Resources Association
• /
• v.56 no.4
• /
• pp.245-259
• /
• 2023

The agricultural reservoir is a critical water supply system in South Korea, providing approximately 60% of the agricultural water demand. However, the reservoir faces several issues that jeopardize its efficient operation and management. To address this issues, we propose a novel deep-learning-based water level recognition model that uses CCTV image data to accurately estimate water levels in agricultural reservoirs. The model consists of three main parts: (1) dataset construction, (2) image segmentation using the U-Net algorithm, and (3) CCTV-based water level recognition using either CNN or ResNet. The model has been applied to two reservoirs G-reservoir and M-reservoir with observed CCTV image and water level time series data. The results show that the performance of the image segmentation model is superior, while the performance of the water level recognition model varies from 50 to 80% depending on water level classification criteria (i.e., classification guideline) and complexity of image data (i.e., variability of the image pixels). The performance of the model can be improved if more numbers of data can be collected.

• Geophysics and Geophysical Exploration
• /
• v.25 no.3
• /
• pp.140-161
• /
• 2022

Seismic data with missing traces are often obtained regularly or irregularly due to environmental and economic constraints in their acquisition. Accordingly, seismic data interpolation is an essential step in seismic data processing. Recently, research activity on machine learning-based seismic data interpolation has been flourishing. In particular, convolutional neural network (CNN) and generative adversarial network (GAN), which are widely used algorithms for super-resolution problem solving in the image processing field, are also used for seismic data interpolation. In this study, CNN-based algorithm, U-Net and GAN-based algorithm, and conditional Wasserstein GAN (cWGAN) were used as seismic data interpolation methods. The results and performances of the methods were evaluated thoroughly to find an optimal interpolation method, which reconstructs with high accuracy missing seismic data. The work process for model training and performance evaluation was divided into two cases (i.e., Cases I and II). In Case I, we trained the model using only the regularly sampled data with 50% missing traces. We evaluated the model performance by applying the trained model to a total of six different test datasets, which consisted of a combination of regular, irregular, and sampling ratios. In Case II, six different models were generated using the training datasets sampled in the same way as the six test datasets. The models were applied to the same test datasets used in Case I to compare the results. We found that cWGAN showed better prediction performance than U-Net with higher PSNR and SSIM. However, cWGAN generated additional noise to the prediction results; thus, an ensemble technique was performed to remove the noise and improve the accuracy. The cWGAN ensemble model removed successfully the noise and showed improved PSNR and SSIM compared with existing individual models.

• Journal of Biomedical Engineering Research
• /
• v.38 no.6
• /
• pp.342-351
• /
• 2017

The neural decoding is a procedure that uses spike trains fired by neurons to estimate features of original stimulus. This is a fundamental step for understanding how neurons talk each other and, ultimately, how brains manage information. In this paper, the strategies of neural decoding are classified into three methodologies: rate decoding, temporal decoding, and population decoding, which are explained. Rate decoding is the firstly used and simplest decoding method in which the stimulus is reconstructed from the numbers of the spike at given time (e. g. spike rates). Since spike number is a discrete number, the spike rate itself is often not continuous and quantized, therefore if the stimulus is not static and simple, rate decoding may not provide good estimation for stimulus. Temporal decoding is the decoding method in which stimulus is reconstructed from the timing information when the spike fires. It can be useful even for rapidly changing stimulus, and our sensory system is believed to have temporal rather than rate decoding strategy. Since the use of large numbers of neurons is one of the operating principles of most nervous systems, population decoding has advantages such as reduction of uncertainty due to neuronal variability and the ability to represent a stimulus attributes simultaneously. Here, in this paper, three different decoding methods are introduced, how the information theory can be used in the neural decoding area is also given, and at the last machinelearning based algorithms for neural decoding are introduced.

• Biomolecules & Therapeutics
• /
• v.20 no.4
• /
• pp.357-370
• /
• 2012

Radiation therapy, the most commonly used for the treatment of brain tumors, has been shown to be of major significance in tumor control and survival rate of brain tumor patients. About 200,000 patients with brain tumor are treated with either partial large field or whole brain radiation every year in the United States. The use of radiation therapy for treatment of brain tumors, however, may lead to devastating functional deficits in brain several months to years after treatment. In particular, whole brain radiation therapy results in a significant reduction in learning and memory in brain tumor patients as long-term consequences of treatment. Although a number of in vitro and in vivo studies have demonstrated the pathogenesis of radiation-mediated brain injury, the cellular and molecular mechanisms by which radiation induces damage to normal tissue in brain remain largely unknown. Therefore, this review focuses on the pathophysiological mechanisms of whole brain radiation-induced cognitive impairment and the identification of novel therapeutic targets. Specifically, we review the current knowledge about the effects of whole brain radiation on pro-oxidative and pro-inflammatory pathways, matrix metalloproteinases (MMPs)/tissue inhibitors of metalloproteinases (TIMPs) system and extracellular matrix (ECM), and physiological angiogenesis in brain. These studies may provide a foundation for defining a new cellular and molecular basis related to the etiology of cognitive impairment that occurs among patients in response to whole brain radiation therapy. It may also lead to new opportunities for therapeutic interventions for brain tumor patients who are undergoing whole brain radiation therapy.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.31 no.4
• /
• pp.351-359
• /
• 2011

In a structure, damage can occur at several scales from micro-cracking to corrosion or loose bolts. This makes the identification of damage difficult with one mode of sensing. Hence, a multi-mode actuated sensing system is proposed based on a self-sensing circuit using a piezoelectric sensor. In the self sensing-based multi-mode actuated sensing, one mode provides a wide frequency-band structural response from the self-sensed impedance measurement and the other mode provides a specific frequency-induced structural wavelet response from the self-sensed guided wave measurement. In this study, an experimental study on the pipeline system is carried out to verify the effectiveness and the robustness of the proposed structural health monitoring approach. Different types of structural damage are artificially inflicted on the pipeline system. To classify the multiple types of structural damage, a supervised learning-based statistical pattern recognition is implemented by composing a two-dimensional space using the damage indices extracted from the impedance and guided wave features. For more systematic damage classification, several control parameters to determine an optimal decision boundary for the supervised learning-based pattern recognition are optimized. Finally, further research issues will be discussed for real-world implementation of the proposed approach.