• Journal of Broadcast Engineering
• /
• v.27 no.2
• /
• pp.228-239
• /
• 2022

In this paper, we propose an unmanned vehicle scratch detection deep learning model for car sharing services. Conventional scratch detection models consist of two steps: 1) a deep learning module for scratch detection of images before and after rental, 2) a manual matching process for finding newly generated scratches. In order to build a fully automatic scratch detection model, we propose a one-step unmanned scratch detection deep learning model. The proposed model is implemented by applying transfer learning and fine-tuning to the deep learning model that detects changes in satellite images. In the proposed car sharing service, specular reflection greatly affects the scratch detection performance since the brightness of the gloss-treated automobile surface is anisotropic and a non-expert user takes a picture with a general camera. In order to reduce detection errors caused by specular reflected light, we propose a preprocessing process for removing specular reflection components. For data taken by mobile phone cameras, the proposed system can provide high matching performance subjectively and objectively. The scores for change detection metrics such as precision, recall, F1, and kappa are 67.90%, 74.56%, 71.08%, and 70.18%, respectively.

• Journal of Korea Entertainment Industry Association
• /
• v.15 no.3
• /
• pp.41-50
• /
• 2021

While artificial intelligence is rapidly replacing human jobs, the art field where human creativity is important is considered an exception. There are currently several AI mastering services in the field of mastering music, a profession at the border between art and technology. In general, the quality of AI mastering is considered to be inferior to the work of a human professional mastering engineer. In this paper, acoustic analysis, listening experiments, and expert interviews were conducted to compare AI and human experts. In the acoustic analysis, In the analysis of audio, there was no significant difference between the results of professional mastering engineers and the results of artificial intelligence. In the listening experiment, the non-musicians could not distinguish between the sound quality of the professional mastering engineer's work and the artificial intelligence work. The group of musicians showed a preference for a specific sound source, but the preference for a specific mastering did not appear significantly. In an expert interview, In expert interviews, respondents answered that there was no significant difference in quality between the two mastering services, and the biggest difference was the communication method between the mastering service provider and the user. In addition, as data increases, it is expected that artificial intelligence mastering will achieve rapid quality improvement and further improvement in communication.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.3
• /
• pp.1100-1118
• /
• 2021

In a wide range of ML applications, the training data contains privacy-sensitive information that should be kept secure. Training the ML systems by privacy-sensitive data makes the ML model inherent to the data. As the structure of the model has been fine-tuned by training data, the model can be abused for accessing the data by the estimation in a reverse process called model inversion attack (MIA). Although, MIA has been applied to shallow neural network models of recognizers in literature and its threat in privacy violation has been approved, in the case of a deep learning (DL) model, its efficiency was under question. It was due to the complexity of a DL model structure, big number of DL model parameters, the huge size of training data, big number of registered users to a DL model and thereof big number of class labels. This research work first analyses the possibility of MIA on a deep learning model of a recognition system, namely a face recognizer. Second, despite the conventional MIA under the white box scenario of having partial access to the users' non-sensitive information in addition to the model structure, the MIA is implemented on a deep face recognition system by just having the model structure and parameters but not any user information. In this aspect, it is under a semi-white box scenario or in other words a gray-box scenario. The experimental results in targeting five registered users of a CNN-based face recognition system approve the possibility of regeneration of users' face images even for a deep model by MIA under a gray box scenario. Although, for some images the evaluation recognition score is low and the generated images are not easily recognizable, but for some other images the score is high and facial features of the targeted identities are observable. The objective and subjective evaluations demonstrate that privacy cyber-attack by MIA on a deep recognition system not only is feasible but also is a serious threat with increasing alert state in the future as there is considerable potential for integration more advanced ML techniques to MIA.

• Journal of the Korea Computer Graphics Society
• /
• v.28 no.3
• /
• pp.1-11
• /
• 2022

Body ownership illusion (BOI), which means 'the illusion that a non-bodily object feels like my body', has been actively studied since the Rubber-hand Illusion showed that a person can feel a rubber hand like one's own hand. It has been proven that BOI can be elicited for virtual bodies in virtual reality. Although many studies have been examined the effect of visuomotor (VM) and visuotactile (VT) stimuli, which are mainly used for the elicitation of BOI, there were very limited studies that delivered both stimuli to the whole body at the same time. In this paper, we investigated how each stimulus affects BOI when delivering VM and VT stimulisimultaneously to the virtual avatar, and examined user experience that appears as presence, emotion, and virtual motion sickness. The results showed that BOI was high when VM is synchronous, but there was no significant difference according to VT levels. In the case of presence, it was confirmed that VT affects only when VM is synchronous, and in the case of emotion, both VM and VT affect valence, and in the case of virtual motion sickness, a statistical difference is not found.These results suggest that overall synchrony of VM is important factor in BOI with virtual avatars, but that the matching VT affects subjective experience such as presence when VM is synchronous.

• Journal of Software Assessment and Valuation
• /
• v.17 no.2
• /
• pp.109-116
• /
• 2021

Traditionally, the engineering field has been dominated by face-to-face education focused on experimental practice, but demand for online learning has soared due to the rapid development of IT technology and Internet communication networks and recent changes in the social environment such as COVID-19. In order for efficient online education to be conducted in the engineering field, where the proportion of experimental practice is relatively high compared to other fields, virtual laboratory practice content that can replace actual experimental practice is very necessary. In this study, we developed a line tracer model and a virtual experimental software to simulate it for efficient online learning of microprocessor applications that are essential not only in the electric and electronic field but also in the overall engineering field where IT convergence takes place. In the developed line tracer model, the user can set various hardware parameter values in the desired form and write the software in assembly language or C language to test the operation on the computer. The developed line tracer virtual experimental software has been used in actual classes to verify its operation, and is expected to be an efficient virtual experimental practice tool in online non-face-to-face classes.

• Journal of Korean Society of Rural Planning
• /
• v.29 no.3
• /
• pp.25-37
• /
• 2023

According to the 「Rural Spatial Reconstruction and Regeneration Support Act」, enacted on March 29, 2024, all local governments are required to establish a 'Rural Spatial Reconstruction and Regeneration Plan' (hereinafter referred to as the 'Rural Spatial Plan'). In order for the 'Rural Spatial Plan' to be appropriately established, this study analyzed the supply and demand of spatial data from the perspective of user stakeholders and derived implications for improving rural spatial planning data utilization. In conclusion, three key recommendations come from this result. Firstly, it is necessary to establish an integrated DB for rural spatial planning data. This can solve the problem of low awareness of scattered data-providing websites, reduce the processing time of non-GIS data, and reduce the time required to acquire data by securing the availability of data search and download. In particular, research should be conducted on the establishment of a spatial analysis simulation system to support stakeholders' decision-making, considering that many stakeholders have difficulty in spatial analysis because spatial analysis techniques were not actively used in rural projects before the implementation of the rural agreement system in 2020. Secondly, research on how to improve data acquisition should be conducted in each data sector. The data sector group with the lowest ease of receiving are 'Local Community Domain', 'Changes in Domestic and International Conditions', and 'Provision and Utilization of Daily Life Services'. Lastly, in-depth research is needed on how to raise each rural spatial planning data supply stakeholder to the position of player. Stakeholders of 'University Institutions' and 'Public Enterprises and Research Institutes' should give those who participate in the formulation of rural spatial plans access to the raw data collected for public work. Stakeholders of 'Private company' need to come up with realistic measures to build a data pool centered on consultative bodies between existing private companies and then prepare a step-by-step strategy to fully open it by participating various stakeholders. In order to induce 'Village Residents and Associations' stakeholders to play a leading role as owners and producers of data, personnel should be trained to collect and record data related to the village. In addition, support measures should be prepared to continue these activities.

• 한국정책학회보
• /
• v.23 no.3
• /
• pp.161-198
• /
• 2014

The various types of improper usages in the process of delivering social services have been increasingly paid to attention in South Korea. This study, relying on empirical data about the various improper behaviors, explore whether the audit activity can reduce the degree of the improper behavior. In order to estimate the impact of the audit, we use the DID(Difference in difference) method, comparing the experimental group with the audit treatment and the control group without it. We control for size(the number of Service personnel and user), types of social service (elderly, disabled, etc.), organizational forms(profit, non-profit), region(metropolitan areas, small-medium cities, rural areas), and the number times of audit (1, 2, and 3 number). Our empirical results show that the audit decreased the ratio of payment violation by about 4.02 percent, the number of violations from providers' improper payment by approximately 5.07 and the number of violations from users' improper payment by approximately 9.59. Further research is required to explore why and how the audit can decrease the improper usage in social service with rigorous theoretical models.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.43 no.3
• /
• pp.353-362
• /
• 2023

In this research, the usage status and characteristics of user big data of Mobike, a dockless bike sharing service introduced in Suwon city, were analyzed, and multiple regression analysis was performed to identify factors influencing the demand for dockless bike sharing service. For analysis, usage data of bike sharing system in Suwon city in 2019 were obtained, and they were organized by areas. As a result of analyzing the characteristics of the influencing factors selected for each area, it was found that the extension of bicycle roads shows high in areas with high demand for bicycles or adjacent areas. Also, the population of 10-30's shows high in areas with high demand for bicycles or adjacent areas. In addition, it was analyzed that the use of bike sharing system is high in areas with high maintenance rate of bicycle roads and large-scale residential and commercial facilities near residential districts and adjacent areas. As a result of the multiple regression analysis, it is analyzed that length of bicycle·pedestrian roads (non-separated), population of 10-30's, number of railway stations, number of schools, number of commercial facilities, number of industrial facilities factors were significant. It is expected that it may be possible to create an environment in which citizens want to use dockless bike sharing service by identifying factors affecting the number of stationless shared bicycles. Also, the results of data analysis are considered to be contributing to policy data to promote the use of dockless bike sharing.

• The Journal of Bigdata
• /
• v.8 no.1
• /
• pp.131-156
• /
• 2023

The traditional retail industry significantly changed over the past decade due to the mobile and online technologies. This change has been accompanied by a shift in consumer behavior regarding purchasing patterns. Despite the rise of online shopping, there are still specific categories of products, such as "Processed food" in Mongolia, for which traditional shopping remains the preferred purchase method. To prepare for the inevitable future of retail businesses, firms need to closely analyze the performance of their offline stores to plan their further actions in a new multi-channel environment. Retailers must integrate diverse channels into their operations to stay relevant and adjust to the shifting market. In this research, we have analyzed the performance data such as sales, profit, and amount of sales of offline stores by using clustering approach. From the clustering, we have found the several distinct insights by comparing the circumstances and performance of retail stores. For the certain retail stores, we have proposed three different strategies: a fulfillment hub store between online and offline channels, an experience store to elongate customers' time on the premises, and a merge between two non-related channels that could complement each other to increase traffic based on the store characteristics. With the proposed strategies, it may enhance the user experience and profit at the same time.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.6D
• /
• pp.1003-1010
• /
• 2006

This study developed a traffic accident information management system based on WebGIS that can process a lot of data for giving effectively diagnosis of traffic accidents in serious damage circumstances by traffic accident. Also, this study presents a way to compose and to convey traffic accident information. In addition, non-spatial attributes as well as spatial attributes about traffic accidents information be integrated and managed by the system. To provide Web service, we developed modules that can supply visually spatial information and traffic accidents data through ASP, Javascript, ArcIMS based on Web and constructed a server. And constructed system include a function that offer the now situation of traffic accident in real time, which supply the statistical data of traffic accident through Web as soon as user entry data in comparison with previous way that preparatory period until traffic accidents data is supplied to peoples had been long. Traffic accidents are analyzed with only nonspatial attribute by simply collecting in the past. However, system constructed by this study offer new function that can grasp visually accident spot circumstance and use detailed content and accurate location data as well as statistical data of traffic accidents. Also, it offer interface that can connect directly with accident charge policeman.