• Smart Structures and Systems
• /
• v.30 no.6
• /
• pp.571-582
• /
• 2022

Optimal life-cycle management is a challenging issue for deteriorating regional bridges. Due to the complexity of regional bridge structural conditions and a large number of inspection and maintenance actions, decision-makers generally choose traditional passive management strategies. They are less efficiency and cost-effectiveness. This paper suggests a deep reinforcement learning framework employing double-deep Q-networks (DDQNs) to improve the life-cycle management of deteriorating regional bridges to tackle these problems. It could produce optimal maintenance plans considering restrictions to maximize maintenance cost-effectiveness to the greatest extent possible. DDQNs method could handle the problem of the overestimation of Q-values in the Nature DQNs. This study also identifies regional bridge deterioration characteristics and the consequence of scheduled maintenance from years of inspection data. To validate the proposed method, a case study containing hundreds of bridges is used to develop optimal life-cycle management strategies. The optimization solutions recommend fewer replacement actions and prefer preventative repair actions when bridges are damaged or are expected to be damaged. By employing the optimal life-cycle regional maintenance strategies, the conditions of bridges can be controlled to a good level. Compared to the nature DQNs, DDQNs offer an optimized scheme containing fewer low-condition bridges and a more costeffective life-cycle management plan.

• Journal of Drive and Control
• /
• v.19 no.3
• /
• pp.9-15
• /
• 2022

This paper proposes a method for constructing and verifying datasets used in deep learning technology, to prevent safety accidents in automated construction machinery or autonomous vehicles. Although open datasets for developing image recognition technologies are challenging to meet requirements desired by users, this study proposes the interface of virtual simulators to facilitate the creation of training datasets desired by users. The pixel-level training image dataset was verified by creating scenarios, including various road types and objects in a virtual environment. Detecting an object from an image may interfere with the accurate path determination due to occlusion areas covered by another object. Thus, we construct a database, for developing an occlusion area detection algorithm in a virtual environment. Additionally, we present the possibility of its use as a deep learning dataset to calculate a grid map, that enables path search considering occlusion areas. Custom datasets are built using the RDBMS system.

• Journal of Information Processing Systems
• /
• v.18 no.1
• /
• pp.75-88
• /
• 2022

With the development of the sharing economy, existing recommender services are changing from user-item recommendations to user-user recommendations. The most important consideration is that all users should have the best possible satisfaction. To achieve this outcome, the matching service adds information between users and items necessary for the existing recommender service and information between users, so higher-level data mining is required. To this end, this paper proposes a user-to-user matching service (UTU-MS) employing the prediction of mutual satisfaction based on learning. Users were divided into consumers and suppliers, and the properties considered for recommendations were set by filtering and weighting. Based on this process, we implemented a convolutional neural network (CNN)-deep neural network (DNN)-based model that can predict each supplier's satisfaction from the consumer perspective and each consumer's satisfaction from the supplier perspective. After deriving the final mutual satisfaction using the predicted satisfaction, a top recommendation list is recommended to all users. The proposed model was applied to match guests with hosts using Airbnb data, which is a representative sharing economy platform. The proposed model is meaningful in that it has been optimized for the sharing economy and recommendations that reflect user-specific priorities.

• The Korean Journal of Franchise Management
• /
• v.6 no.2
• /
• pp.51-66
• /
• 2015

As competition for better service between corporations is getting fierce, many efforts to improve service quality are being made endlessly. The quality of service is directly associated with customer satisfaction and the employee performance of emotional labor is a key factor to attain a high level of customer review and better corporation image. This study examines an influence of emotional labor on job stress and customer orientation in the context of a service franchise firm. The results are as follow. First, deep acting of emotional labor has a negative influence on job stress, and surface acting of emotional labor has a negative influence on customer orientation. Also, job stress has a negative influence on customer orientation. Second, deep acting of emotional labor does not have a positive influence on customer orientation. Third, surface acting of emotional labor does not have a positive influence on job stress. The findings of this study show that deep acting of service based on an employee emotion can produce the employee's better service attitude by decreasing employees' job stress, but standardized surface acting of service cannot. Therefore, franchisor needs to use employees' deep acting to improve the franchisee service quality.

• Restorative Dentistry and Endodontics
• /
• v.47 no.4
• /
• pp.45.1-45.10
• /
• 2022

Objectives: This study evaluated the relationship between the battery charge level and irradiance of light-emitting diode (LED) light-curing units (LCUs) and how these variables influence the Vickers hardness number (VHN) of a bulk-fill resin. Materials and Methods: Four LCUs were evaluated: Radii Plus (SDI), Radii-cal (SDI), Elipar Deep Cure (Filtek Bulk Fill, 3M Oral Care), and Poly Wireless (Kavo Kerr). Irradiance was measured using a radiometer every ten 20-second activations until the battery was discharged. Disks (4 mm thick) of a bulk-fill resin (Filtek Bulk Fill, 3M Oral Care) were prepared, and the VHN was determined on the top and bottom surfaces when light-cured with the LCUs with battery levels at 100%, 50% and 10%. Data were analyzed by 2-way analysis of variance, the Tukey's test, and Pearson correlations (α = 5%). Results: Elipar Deep Cure and Poly Wireless showed significant differences between the irradiance when the battery was fully charged versus discharged (10% battery level). Significant differences in irradiance were detected among all LCUs, within each battery condition tested. Hardness ratios below 80% were obtained for Radii-cal (10% battery level) and for Poly Wireless (50% and 10% battery levels). The battery level showed moderate and strong, but non-significant, positive correlations with the VHN and irradiance. Conclusions: Although the irradiance was different among LCUs, it decreased in half of the devices along with a reduction in battery level. In addition, the composite resin effectiveness of curing, measured by the hardness ratio, was reduced when the LCUs' battery was discharged.

• Economic and Environmental Geology
• /
• v.28 no.3
• /
• pp.279-286
• /
• 1995

The amount of the high level radioactive wastes in Korea will be increased up to 14,297 MTU about 2010 year. Most of countries adopt the concept of deep burial repository in high level radioactive waste disposal. Because the high level radioactive wastes are very toxic in biosphere and to human, the data verifing its never return to the biosphere are requisite for the disposal. Presently, the evaluating techniques for the high level radioactive waste disposal are not fully developed. Therefore, in order to dispose the high level radioactive wastes in proper time the R & D of it is urged in our country. The R & D and/or the international joint research programme for the disposal of high level wastes have already been proceeded. In our country no plan for its disposal has been prepared. It is the time that the direction of the R & D is to be discused seriously. The R & D for the disposal of high level radioactive wastes in Korea is believed to be focused on developing the pecular techniques such as in situ characteristics of groundwater flowage, and change of properties of in situ rock mass at thermal effects.

• KIISE Transactions on Computing Practices
• /
• v.23 no.4
• /
• pp.244-249
• /
• 2017

Steganalysis is to detect information hidden by steganography inside general data such as images. There are stegoanalysis techniques that use machine learning (ML). Existing ML approaches to steganalysis are based on extracting features from stego images and modeling them. Recently deep learning-based methodologies have shown significant improvements in detection accuracy. However, all the existing methods, including deep learning-based ones, have a critical limitation in that they can only detect stego images that are created by a specific steganography method. In this paper, we propose a generalized steganalysis method that can model multiple types of stego images using deep learning. Through various experiments, we confirm the effectiveness of our approach and envision directions for future research. In particular, we show that our method can detect each type of steganography with the same level of accuracy as that of a steganalysis method dedicated to that type of steganography, thereby demonstrating the general applicability of our approach to multiple types of stego images.

• Korean Journal of Remote Sensing
• /
• v.33 no.4
• /
• pp.423-436
• /
• 2017

This paper presents a deep learning-based road segmentation framework from very high-resolution orthophotos. The proposed method uses Deep Convolutional Autoencoders for end-to-end mapping of orthophotos to road segmentations. In addition, a set of post-processing steps were applied to make the model outputs GIS-ready data that could be useful for various applications. The optimization of the model's parameters is explained which was conducted via grid search method. The model was trained and implemented in Keras, a high-level deep learning framework run on top of Tensorflow. The results show that the proposed model with the best-obtained hyperparameters could segment road objects from orthophotos at an average accuracy of 88.5%. The results of optimization revealed that the best optimization algorithm and activation function for the studied task are Stochastic Gradient Descent (SGD) and Exponential Linear Unit (ELU), respectively. In addition, the best numbers of convolutional filters were found to be 8 for the first and second layers and 128 for the third and fourth layers of the proposed network architecture. Moreover, the analysis on the time complexity of the model showed that the model could be trained in 4 hours and 50 minutes on 1024 high-resolution images of size $106{\times}106pixels$, and segment road objects from similar size and resolution images in around 14 minutes. The results show that the deep learning models such as Convolutional Autoencoders could be a best alternative to traditional machine learning models for road segmentation from aerial photographs.

• Ocean and Polar Research
• /
• v.32 no.1
• /
• pp.15-22
• /
• 2010

Calcium carbonate ($CaCO_3$) content was measured from 3 box core (BC060301, BC060303, BC070301) sediments, in addition to pilot core (PC313) sediments, from deep waters within the Western Pacific Ocean. At the two collection sites (BC060301, PC313) located close to the equator, downcore variation exhibited low $CaCO_3$ content during the interglacial period and high $CaCO_3$ content during the glacial period. Variation of coarse fraction (>$63\;{\mu}m$) content also followed changes in $CaCO_3$ content, indicating that dissolution effect of bottom water decreased during the glacial period. Such variation pattern is typical of the Pacific Ocean. However, downcore variation at the two collection sites (BC060303, BC070301) in the Philippine Sea contrasted the trend of the previous two cores (i.e., high $CaCO_3$ content during the interglacial period and low during the glacial period). This pattern is typical of the Atlantic Ocean. Such results may be attributed to the increasing dilution effect, initiated possibly by the increased transportation of terrigenous materials from nearby continent and archipelago during the glacial period when sea level was low. Alternatively, it is possible that the non-carbonate biogenic particles may have been responsible for dilution. Because of these uncertainties, the record of $CaCO_3$ variation in the deep Western Pacific Ocean is not regionally consistent.

• Steel and Composite Structures
• /
• v.17 no.4
• /
• pp.471-495
• /
• 2014

The responses of steel buildings with perimeter moment resisting frames (PMRF) with medium size columns (W14) are estimated and compared with those of buildings with deep columns (W27), which are selected according to two criteria: equivalent resistance and equivalent weight. It is shown that buildings with W27 columns have no problems of lateral torsional, local or shear buckling in panel zone. Whether the response is larger for W14 or W27 columns, depends on the level of deformation, the response parameter and the structural modeling under consideration. Modeling buildings as two-dimensional structures result in an overestimation of the response. For multiple response parameters, the W14 columns produce larger responses for elastic behavior. The axial load on columns may be significantly larger for the buildings with W14 columns. The interstory displacements are always larger for W14 columns, particularly for equivalent weight and plane models, implying that using deep columns helps to reduce interstory displacements. This is particularly important for tall buildings where the design is usually controlled by the drift limit state. The interstory shears in interior gravity frames (GF) are significantly reduced when deep columns are used. This helps to counteract the no conservative effect that results in design practice, when lateral seismic loads are not considered in GF of steel buildings with PMRF. Thus, the behavior of steel buildings with deep columns, in general, may be superior to that of buildings with medium columns, using less weight and representing, therefore, a lower cost.