• Journal of the Korean GEO-environmental Society
• /
• v.22 no.10
• /
• pp.13-19
• /
• 2021

Cracks in asphalt pavement occur due to changes in weather or impact from vehicles, and if cracks are left unattended, the life of the pavement may be shortened, and various accidents may occur. Therefore, studies have been conducted to detect cracks through images in order to quickly detect cracks in the asphalt pavement automatically and perform maintenance activity. Recent studies adopt machine-learning models for detecting cracks in asphalt road pavement using a Convolutional Neural Network. However, their practical use is limited because they require high-performance computing power. Therefore, this paper proposes a framework for detecting cracks in asphalt road pavement by applying a small deep learning model applicable to mobile devices. The small deep learning model proposed through the case study was compared with general deep learning models, and although it was a model with relatively few parameters, it showed similar performance to general deep learning models. The developed model is expected to be embedded and used in mobile devices or IoT for crack detection in asphalt pavement.

• Journal of the Korea Computer Graphics Society
• /
• v.25 no.1
• /
• pp.23-32
• /
• 2019

Cybersickness is a symptom of dizziness that occurs while experiencing Virtual Reality (VR) technology and it is presumed to occur mainly by crosstalk between the sensory and cognitive systems. However, since the sensory and cognitive systems cannot be measured objectively, it is difficult to measure cybersickness. Therefore, methodologies for measuring cybersickness have been studied in various ways. Traditional studies have collected answers to questionnaires or analyzed EEG data using machine learning algorithms. However, the system relying on the questionnaires lacks objectivity, and it is difficult to obtain highly accurate measurements with the machine learning algorithms. In this work, we apply Deep Neural Network (DNN) deep learning algorithm for objective cybersickness measurement from EEG data. We also propose a data preprocessing for learning and network structures allowing us to achieve high performance when learning EEG data with the deep learning algorithms. Our approach provides cybersickness measurement with an accuracy up to 98.88%. Besides, we analyze video characteristics where cybersickness occurs by examining the video segments causing cybersickness in the experiments. We discover that cybersickness happens even in unusually persistent changes in the darkness such as the light in a room keeps switching on and off.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.2
• /
• pp.20-25
• /
• 2021

The target-object classification method was implemented using a deep-learning-based detection model in real-time images. The object detection model was a deep-learning-based detection model that allowed extensive data collection and machine learning processes to classify similar target-objects. The recognition model was implemented by changing the processing structure of the detection model and combining developed the vision-processing module. To classify the target-objects, the identity and similarity were defined and applied to the detection model. The use of the recognition model in industry was also considered by verifying the effectiveness of the recognition model using the real-time images of an actual soccer game. The detection model and the newly constructed recognition model were compared and verified using real-time images. Furthermore, research was conducted to optimize the recognition model in a real-time environment.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.1
• /
• pp.50-55
• /
• 2021

Multi-path system is a system in which utilizes various networks simultaneously. It is expected that multi-path system can enhance communication speed, reliability, security of network. In this paper, we focus on path selection in multi-path system. To select optimal path, we propose deep reinforcement learning algorithm which is rewarded by the round-trip-time (RTT) of each networks. Unlike multi-armed bandit model, deep Q learning is applied to consider rapidly changing situations. Due to the delay of RTT data, we also suggest compensation algorithm of the delayed reward. Moreover, we implement testbed learning server to evaluate the performance of proposed algorithm. The learning server contains distributed database and tensorflow module to efficiently operate deep learning algorithm. By means of simulation, we showed that the proposed algorithm has better performance than lowest RTT about 20%.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1007-1014
• /
• 2022

Compact Advanced Satellite 500 (CAS500) can be used for various purposes, including vegetation, forestry, and agriculture fields. It is expected that it will be possible to acquire satellite images of various areas quickly. In order to use satellite images acquired through CAS500 in the agricultural field, it is necessary to develop a satellite image-based extraction technique for crop-cultivated areas.In particular, as research in the field of deep learning has become active in recent years, research on developing a deep learning model for extracting crop cultivation areas and generating training data is necessary. This manuscript classified the onion and garlic cultivation areas in Hapcheon-gun using PlanetScope satellite images and farm maps. In particular, for effective model learning, the model performance was analyzed according to the proportion of crop-cultivated areas. For the deep learning model used in the experiment, Fully Convolutional Densely Connected Convolutional Network (FC-DenseNet) was reconstructed to fit the purpose of crop cultivation area classification and utilized. As a result of the experiment, the ratio of crop cultivation areas in the training data affected the performance of the deep learning model.

• Journal of the Korea Computer Graphics Society
• /
• v.28 no.3
• /
• pp.45-54
• /
• 2022

Image depth estimation is a technology that is the basis of various image analysis. As analysis methods using deep learning models emerge, studies using deep learning in image depth estimation are being actively conducted. Currently, most deep learning-based depth estimation models are being trained with clean and ideal images. However, due to the lack of data on adverse conditions such as haze or fog, the depth estimation may not work well in such an environment. It is hard to sufficiently secure an image in these environments, and in particular, obtaining non-homogeneous haze data is a very difficult problem. In order to solve this problem, in this study, we propose a method of synthesizing non-homogeneous haze images and a learning method for a monocular depth estimation deep learning model using this method. Considering that haze mainly occurs outdoors, datasets mainly containing outdoor images are constructed. Experiment results show that the model with the proposed method is good at estimating depth in both synthesized and real haze data.

• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.311-323
• /
• 2023

Ever since ancient times, earthquakes have been a major threat to the civil infrastructures and the safety of human beings. The majority of casualties in earthquake disasters are caused by the damaged civil infrastructures but not by the earthquake itself. Therefore, the efficient and accurate post-earthquake assessment of the conditions of structural damage has been an urgent need for human society. Traditional ways for post-earthquake structural assessment rely heavily on field investigation by experienced experts, yet, it is inevitably subjective and inefficient. Structural response data are also applied to assess the damage; however, it requires mounted sensor networks in advance and it is not intuitional. As many types of damaged states of structures are visible, computer vision-based post-earthquake structural assessment has attracted great attention among the engineers and scholars. With the development of image acquisition sensors, computing resources and deep learning algorithms, deep learning-based post-earthquake structural assessment has gradually shown potential in dealing with image acquisition and processing tasks. This paper comprehensively reviews the state-of-the-art studies of deep learning-based post-earthquake structural assessment in recent years. The conventional way of image processing and machine learning-based structural assessment are presented briefly. The workflow of the methodology for computer vision and deep learning-based post-earthquake structural assessment was introduced. Then, applications of assessment for multiple civil infrastructures are presented in detail. Finally, the challenges of current studies are summarized for reference in future works to improve the efficiency, robustness and accuracy in this field.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.131-131
• /
• 2021

Deep learning models, especially those based on long short-term memory (LSTM), have presented their superiority in addressing time series data issues recently. This study aims to comprehensively evaluate the performance of deep learning models that belong to the supervised learning category in streamflow prediction. Therefore, six deep learning models-standard LSTM, standard gated recurrent unit (GRU), stacked LSTM, bidirectional LSTM (BiLSTM), feed-forward neural network (FFNN), and convolutional neural network (CNN) models-were of interest in this study. The Red River system, one of the largest river basins in Vietnam, was adopted as a case study. In addition, deep learning models were designed to forecast flowrate for one- and two-day ahead at Son Tay hydrological station on the Red River using a series of observed flowrate data at seven hydrological stations on three major river branches of the Red River system-Thao River, Da River, and Lo River-as the input data for training, validation, and testing. The comparison results have indicated that the four LSTM-based models exhibit significantly better performance and maintain stability than the FFNN and CNN models. Moreover, LSTM-based models may reach impressive predictions even in the presence of upstream reservoirs and dams. In the case of the stacked LSTM and BiLSTM models, the complexity of these models is not accompanied by performance improvement because their respective performance is not higher than the two standard models (LSTM and GRU). As a result, we realized that in the context of hydrological forecasting problems, simple architectural models such as LSTM and GRU (with one hidden layer) are sufficient to produce highly reliable forecasts while minimizing computation time because of the sequential data nature.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.1
• /
• pp.435-441
• /
• 2024

In order for the large amount of collected data sets to be used as deep learning training data, sensitive personal information such as resident registration number and disease information must be changed or encrypted to prevent it from being exposed to hackers, and the data must be reconstructed to match the structure of the built deep learning model. Currently, these tasks are performed manually by experts, which takes a lot of time and money. To solve these problems, this paper proposes a technique that can automatically perform data processing tasks to protect personal information during the deep learning process. In the proposed technique, privacy protection tasks are performed based on data generalization and data reconstruction tasks are performed using circular queues. To verify the validity of the proposed technique, it was directly implemented using C language. As a result of the verification, it was confirmed that data generalization was performed normally and data reconstruction suitable for the deep learning model was performed properly.

• Korean Journal of Artificial Intelligence
• /
• v.5 no.1
• /
• pp.1-9
• /
• 2017

Recently, the Q-Learning algorithm, which is one kind of reinforcement learning, is mainly used to implement artificial intelligence system in combination with deep learning. Many research is going on to improve the performance of Q-Learning. Therefore, purpose of theory try to improve the performance of Q-Learning algorithm. This Theory apply Cross Entropy Error to the loss function of Q-Learning algorithm. Since the mean squared error used in Q-Learning is difficult to measure the exact error rate, the Cross Entropy Error, known to be highly accurate, is applied to the loss function. Experimental results show that the success rate of the Mean Squared Error used in the existing reinforcement learning was about 12% and the Cross Entropy Error used in the deep learning was about 36%. The success rate was shown.