• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.40 no.3
• /
• pp.303-314
• /
• 2020

The hydrological characteristics of watersheds and hydraulic systems of urban and river floods are highly nonlinear and contain uncertain variables. Therefore, the predicted time series of rainfall-runoff data in flood analysis is not suitable for existing neural networks. To overcome the challenge of prediction, a NARX (Nonlinear Autoregressive Exogenous Model), which is a kind of recurrent dynamic neural network that maximizes the learning ability of a neural network, was applied to forecast a flood in real-time. At the same time, NARX has the characteristics of a time-delay neural network. In this study, a hydrological model was constructed for the Taehwa river basin, and the NARX time-delay parameter was adjusted 10 to 120 minutes. As a result, we found that precise prediction is possible as the time-delay parameter was increased by confirming that the NSE increased from 0.530 to 0.988 and the RMSE decreased from 379.9 ㎥/s to 16.1 ㎥/s. The machine learning technique with NARX will contribute to the accurate prediction of flow rate with an unexpected extreme flood condition.

• Journal of Korea Technology Innovation Society
• /
• v.17 no.2
• /
• pp.309-333
• /
• 2014

To meet the challenge of new type of innovation activities requires us to understand the social context of innovation and the potential needs of innovation users and, based on this, to co-construct technology and society simultaneously. Effective 'demand articulation' activities such as the understanding and utilization of user experiences and socio-technical planning are prerequisites for carrying out post-catch up innovations shaping new trajectories and contributing to solving social problems. Living Lab has recently been emerging particularly in Europe as an 'user-driven innovation model', in which users are active participants in innovation activities. The purpose of this study is to contribute to a theoretical discussion of Living Lab as an user-driven innovation model, to make a brief review of cases of Living Lab and to explore Living Lab's applicability in the Korean context. Living Lab is an open innovation model, in which end suers actively participate in innovation processes in a particular geographical space or region and would be able to solve specific problems of that space or region. In that sense, Living Lab would be able to strengthen the problem-solving capabilities of local communities and to become a pioneer in inducing and realizing a new socio-technical system. Furthermore, Living Lab could become an innovative policy tool reflecting recent major changes in innovation policy paradigms such as post-catch up innovation, demand-oriented innovation, regional innovation, societal innovation, innovation eco-system and socio-technical system transition, and thus make a contribution to exploring a new way of bringing about changes in the Korean society.

• Journal of Internet Computing and Services
• /
• v.19 no.5
• /
• pp.43-54
• /
• 2018

Images taken in haze weather are characteristic of low contrast and poor visibility. The process of reconstructing clear-weather image from a hazy image is called dehazing. The main challenge of image dehazing is to estimate the transmission map or depth map for an input hazy image. In this paper, we propose a single image dehazing method by utilizing the Generative Adversarial Network(GAN) for accurate depth map estimation. The proposed GAN model is trained to learn a nonlinear mapping between the input hazy image and corresponding depth map. With the trained model, first the depth map of the input hazy image is estimated and used to compute the transmission map. Then a guided filter is utilized to preserve the important edge information of the hazy image, thus obtaining a refined transmission map. Finally, the haze-free image is recovered via atmospheric scattering model. Although the proposed GAN model is trained on synthetic indoor images, it can be applied to real hazy images. The experimental results demonstrate that the proposed method achieves superior dehazing results against the state-of-the-art algorithms on both the real hazy images and the synthetic hazy images, in terms of quantitative performance and visual performance.

• Korean Security Journal
• /
• no.17
• /
• pp.361-381
• /
• 2008

Quality rating of machine security systems is difficult because both tangible and intangible services are included. However, still, the research template applied the SERVQUAL model with the intention of confirming machine security systems' service quality formation and experimentally inspecting the relationship between service quality and customer satisfaction. Therefore, the following highlights the experimental research outcomes and their implications for small-scale businesses utilizing machine security systems in the Daegu region. First, after observing whether the determining factors constitute service quality, four components were found to have significant influence on customer satisfaction. Additionally, in observing any differences in their influences, the following in order were observed as having influence on customer satisfaction: empathy, assurance reliability, responsiveness, and tangibility. Moreover, though companies‘ newest facilities and equipment are important, it can be concluded that a company employees’ prudent consideration, individual interest, reliability, and assurance for the customer carry greater importance. Secondly, though we intended to survey machine security systems by employing the SERVQUAL model, determinant factor analysis results found applying SERVQUAL model in its original state a challenge. According to results from determinant factor analysis, the basis for forming service quality is determined by assurance reliability, empathy, tangibility, and responsiveness. Furthermore, in future research, while more accurately distinguishing between assurance and reliability, a more appropriate model must also be considered for modification in domestic machine security system industry‘s service quality evaluation.

• Journal of Gastric Cancer
• /
• v.24 no.3
• /
• pp.246-256
• /
• 2024

Purpose: Peritoneal carcinomatosis (PC) presents a major challenge in the treatment of late-stage, solid tumors, with traditional therapies limited by poor drug penetration. We evaluated a novel hyperthermic pressurized intraperitoneal aerosol chemotherapy (HPIPAC) system using a human abdominal cavity model for its efficacy against AGS gastric cancer cells. Materials and Methods: A model simulating the human abdominal cavity and AGS gastric cancer cell line cultured dishes were used to assess the efficacy of the HPIPAC system. Cell viability was measured to evaluate the impact of HPIPAC under 6 different conditions: heat alone, PIPAC with paclitaxel (PTX), PTX alone, normal saline (NS) alone, heat with NS, and HPIPAC with PTX. Results: Results showed a significant reduction in cell viability with HPIPAC combined with PTX, indicating enhanced cytotoxic effects. Immediately after treatment, the average cell viability was 66.6%, which decreased to 49.2% after 48 hours and to a further 19.6% after 120 hours of incubation, demonstrating the sustained efficacy of the treatment. In contrast, control groups exhibited a recovery in cell viability; heat alone showed cell viability increasing from 90.8% to 94.4%, PIPAC with PTX from 82.7% to 89.7%, PTX only from 73.3% to 74.8%, NS only from 90.9% to 98.3%, and heat with NS from 74.4% to 84.7%. Conclusions: The HPIPAC system with PTX exhibits a promising approach in the treatment of PC in gastric cancer, significantly reducing cell viability. Despite certain limitations, this study highlights the system's potential to enhance treatment outcomes. Future efforts should focus on refining HPIPAC and validating its effectiveness in clinical settings.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.43 no.6 s.312
• /
• pp.28-35
• /
• 2006

Statistical models of shape variability based on active shape models (ASMs) have been successfully utilized to perform segmentation and recognition tasks in two-dimensional (2D) images. Three-dimensional (3D) model-based approaches are more promising than 2D approaches since they can bring in more realistic shape constraints for recognizing and delineating the object boundary. For 3D model-based approaches, however, building the 3D shape model from a training set of segmented instances of an object is a major challenge and currently it remains an open problem in building the 3D shape model, one essential step is to generate a point distribution model (PDM). Corresponding landmarks must be selected in all1 training shapes for generating PDM, and manual determination of landmark correspondences is very time-consuming, tedious, and error-prone. In this paper, we propose a novel automatic method for generating 3D statistical shape models. Given a set of training 3D shapes, we generate a 3D model by 1) building the mean shape fro]n the distance transform of the training shapes, 2) utilizing a tetrahedron method for automatically selecting landmarks on the mean shape, and 3) subsequently propagating these landmarks to each training shape via a distance labeling method. In this paper, we investigate the accuracy and compactness of the 3D model for the human liver built from 50 segmented individual CT data sets. The proposed method is very general without such assumptions and can be applied to other data sets.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.89-106
• /
• 2022

Recently, the amount of video data collected from smartphones, CCTVs, black boxes, and high-definition cameras has increased rapidly. According to the increasing video data, the requirements for analysis and utilization are increasing. Due to the lack of skilled manpower to analyze videos in many industries, machine learning and artificial intelligence are actively used to assist manpower. In this situation, the demand for various computer vision technologies such as object detection and tracking, action detection, emotion detection, and Re-ID also increased rapidly. However, the object detection and tracking technology has many difficulties that degrade performance, such as re-appearance after the object's departure from the video recording location, and occlusion. Accordingly, action and emotion detection models based on object detection and tracking models also have difficulties in extracting data for each object. In addition, deep learning architectures consist of various models suffer from performance degradation due to bottlenects and lack of optimization. In this study, we propose an video analysis system consists of YOLOv5 based DeepSORT object tracking model, SlowFast based action recognition model, Torchreid based Re-ID model, and AWS Rekognition which is emotion recognition service. Proposed model uses single-linkage hierarchical clustering based Re-ID and some processing method which maximize hardware throughput. It has higher accuracy than the performance of the re-identification model using simple metrics, near real-time processing performance, and prevents tracking failure due to object departure and re-emergence, occlusion, etc. By continuously linking the action and facial emotion detection results of each object to the same object, it is possible to efficiently analyze videos. The re-identification model extracts a feature vector from the bounding box of object image detected by the object tracking model for each frame, and applies the single-linkage hierarchical clustering from the past frame using the extracted feature vectors to identify the same object that failed to track. Through the above process, it is possible to re-track the same object that has failed to tracking in the case of re-appearance or occlusion after leaving the video location. As a result, action and facial emotion detection results of the newly recognized object due to the tracking fails can be linked to those of the object that appeared in the past. On the other hand, as a way to improve processing performance, we introduce Bounding Box Queue by Object and Feature Queue method that can reduce RAM memory requirements while maximizing GPU memory throughput. Also we introduce the IoF(Intersection over Face) algorithm that allows facial emotion recognized through AWS Rekognition to be linked with object tracking information. The academic significance of this study is that the two-stage re-identification model can have real-time performance even in a high-cost environment that performs action and facial emotion detection according to processing techniques without reducing the accuracy by using simple metrics to achieve real-time performance. The practical implication of this study is that in various industrial fields that require action and facial emotion detection but have many difficulties due to the fails in object tracking can analyze videos effectively through proposed model. Proposed model which has high accuracy of retrace and processing performance can be used in various fields such as intelligent monitoring, observation services and behavioral or psychological analysis services where the integration of tracking information and extracted metadata creates greate industrial and business value. In the future, in order to measure the object tracking performance more precisely, there is a need to conduct an experiment using the MOT Challenge dataset, which is data used by many international conferences. We will investigate the problem that the IoF algorithm cannot solve to develop an additional complementary algorithm. In addition, we plan to conduct additional research to apply this model to various fields' dataset related to intelligent video analysis.

• Information Systems Review
• /
• v.12 no.1
• /
• pp.209-227
• /
• 2010

One type of innovative multimedia platform environments is mobile data services (MDS), exemplified by Nate, Show, and OZ. In the context of MDS, enhancing user's continuance intention is a significant challenge to the continuing growth and long-term viability of MDS. Because the cost of using MDS is borne mainly by users, they are likely to evaluate it based on perceptions of what is received and what is given. This study identifies perceived usefulness and perceived enjoyment as the 'get'components, and perceived fee and perceived anxiety as the 'give' components. To understand the role of get and give components in the MDS post-adoption environment, this study incorporates these components into expectation confirmation model. We collected data from 204 users who had direct experiences with MDS within recent 3 months. The data was analyzed by employing PLS (partial least squares). Theoretical and practical implications of our findings are discussed.

• The Journal of Industrial Distribution & Business
• /
• v.11 no.2
• /
• pp.41-48
• /
• 2020

Purpose: This study is to examine the intention of the elderly who live alone in the customized AI speaker for the elderly living alone to improve the quality of life service for the elderly living alone in the smart city environment. Based on the quality of life model of the elderly, this study is applied to the technology acceptance model to investigate the relationship between perceived usefulness and ease of use on the sustained use intention. Research design, data and methodology: Residents in Suwon, Gyeonggi-do, selected as candidate local governments for the Smart City Challenge Project of the Ministry of Land, Infrastructure and Transport in June 2019 to measure the perceived technology acceptance of potential users for the AI technology for the elderly living alone as part of the smart city technology. In order to evaluate the intention of using AI speaker, which is the target system of this study, a video of a chatbot using experience of elderly people living alone was produced. Results: First of all, in order for the elderly living alone to have an attitude to use AI-based speakers, there should be a perceived usefulness of the quality of life of the elderly. However, ease of use did not show any significant causal relationship to attitude toward use. In addition, the attitude toward use weakly influenced the intention to use. In other words, elderly people living alone were not likely to have a significant effect on their attitude toward use. However, feeling that AI speakers are easy to use will help to improve the quality of life, which in turn led to the attitude toward using AI speakers, which could lead to indirect effects. Finally, the perceived usefulness of quality of life was found to have a weak effect on direct use intentions. Conclusions: This study conducted a study on the technology acceptance of service environment to improve the quality of life for the specific user group who live alone in the smart seat environment. In this study, we examined the effects of AI speaker on the elderly living alone to improve the quality of life for the elderly living alone.

• Journal of Distribution Science
• /
• v.9 no.4
• /
• pp.29-42
• /
• 2011

With the many advantages of the internet, online shopping has become one of the fastest growing types of retail businesses. However, internet-based firms are much more firmly required to retain existing customers rather than secure new ones, and to make them revisit the site by strengthening trust and loyalty, thereby improving profits and outrivaling competitors. Commitment is an essential part of successful long-term relationships between buyers and sellers. Although commitments by both parties in an exchange can provide the foundation for the development of relational social norms, disproportionate commitments can lead to opportunism by the less committed partner. Moreover, flow, which is characterized by intense concentration and enjoyment, was found to be significantly linked with exploratory use behavior, which in turn was linked to the extent of computer use. The level of flow was, itself, determined by the individual's sense of being in control, and the level of challenge perceived in maneuvering a website. Website attractiveness goes hand in hand with the attractiveness of an internet shopping mall, and it can be conceptualized as the persuasive effectiveness of a message by the use of familiarity, favor, similarity, etc. It occurs when information receivers try to achieve self-satisfaction when they actually or emotionally identify themselves with an information source. This study investigates the relationship between the perceived system characteristics of an internet shopping mall and the loyalty of online consumers, and it examines how perceived website attractiveness and flow play mediating roles between the perceived system characteristics of an internet shopping mall and the affective commitment in the context of a clothes internet shopping mall. For these purposes, a structural model comprising several variables was developed. That model was tested with an analysis of moment structure (AMOS) using data from respondents who had purchased clothing through the internet during the past three months. In this model, the perceived system characteristics of an internet shopping mall, such as familiarity, reputation, uniqueness, positive emotions, self-efficacy, and interactivity, were proposed to affect the website's attractiveness and flow, and lead to a higher affective commitment over time. Thus, the perceived website attractiveness and flow were proposed as core mediating variables between perceived system characteristics and affective commitment. The results of a reliability test using Cronbach's Alpha, and a confirmatory factor analysis warranted using unidimensionality for the measures for each construct. In addition, the nomological validity of the measures was warranted from the results of a correlation analysis. The results of empirical analyses indicated that systematic attributes resulting in website attractiveness and user's characteristics, thereby triggering customers' flow, play a crucial role in inducing customers' affective commitment, and a user's characteristics are twice as important as systematic attributes in this study. Moreover, familiarity, reputation, and uniqueness all have a significant effect on website attractiveness, and the research showed that uniqueness took the first place, and that familiarity and reputation followed in order of magnitude. The fact that reputation was not the most important factor that affects the attractiveness of an internet shopping mall, with uniqueness or familiarity having a greater impact, suggests much deeper implications. Finally, positive emotion, self-efficacy, and interactivity all have a significant effect on customers' flow. In particular, the fact that positive emotion, compared to self-efficacy or interactivity, has much more impact on flow is very suggestive.