• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권7호
• /
• pp.2547-2567
• /
• 2021

It has been widely acknowledged that occlusion impairments adversely distress many face recognition algorithms' performance. Therefore, it is crucial to solving the problem of face image occlusion in face recognition. To solve the image occlusion problem in face recognition, this paper aims to automatically de-occlude the human face majority or discriminative regions to improve face recognition performance. To achieve this, we decompose the generative process into two key stages and employ a separate generative adversarial network (GAN)-based network in both stages. The first stage generates an initial coarse face image without an occlusion mask. The second stage refines the result from the first stage by forcing it closer to real face images or ground truth. To increase the performance and minimize the artifacts in the generated result, a new refine loss (e.g., reconstruction loss, perceptual loss, and adversarial loss) is used to determine all differences between the generated de-occluded face image and ground truth. Furthermore, we build occluded face images and corresponding occlusion-free face images dataset. We trained our model on this new dataset and later tested it on real-world face images. The experiment results (qualitative and quantitative) and the comparative study confirm the robustness and effectiveness of the proposed work in removing challenging occlusion masks with various structures, sizes, shapes, types, and positions.

• The Journal of Asian Finance, Economics and Business
• /
• 제8권7호
• /
• pp.235-244
• /
• 2021

This study aims to determine the behavior of consumer interest in sustainable fashion products in Indonesia and the factors that influence it. Personal values, Green Trust, and attitude toward green products are studied to get a proper picture of the ethical behavior of consumers. The method used is the Structural Equation Model (SEM) technique using AMOS 23 software. A sample of 103 participants used the convenience sampling technique. The study results show that green trust and attitude toward green products successfully mediate personal value and green purchase intention on ethical fashion products in Indonesia. In this study, it was found that personal values cannot directly influence the purchase intention of sustainable fashion products. The influence of Personal Value must first be mediated by green trust or attitude toward green products to further influence green purchase intention of ethical fashion products in Indonesia. This is not in line with previous studies. Indicators of personal values such as self-transcendence, self-enhancement, conservation, and openness directly cannot predict the behavior of purchase intentions for sustainable fashion products in Indonesia. From the descriptive conclusion of the data obtained, there are various types of meanings of individual values, and fashion consumers in Indonesia are more affected by contextual factors.

• 유통과학연구
• /
• 제19권3호
• /
• pp.61-70
• /
• 2021

Purpose: This paper reviews the attitudes of consumers related to the consumption crisis response strategy (i.e., defensive vs. receptive) that companies implement during crises. Research design, data, and methodology: We discuss the interaction between the crisis response strategy and the consumption crisis type (i.e., corporate ability vs. corporate social responsibility). We used SAS ver. 9.4 software to analyze the results. We applied a 2 × 2 intergroup experimental design to our sample of subjects, who were undergraduate and graduate students at a university in Seoul, South Korea. The three experimental variables considered were the entity's risk response strategy, the crisis type, and public relations strategy. The experiments were conducted by presenting a hypothetical scenario to eight groups. Prior to this experiment, five preliminary surveys were conducted to determine the three variables just mentioned. Preliminary surveys were conducted on the basis of these criteria. For eight selected product lines, 320 undergraduates were required to enter the product lines that are frequently used in the assessment center up to the fourth priority. Results: Analysis of variance revealed that experiments related to crisis response strategy, type of enterprise crisis, and type of public relations message were successful. Conclusions: Our results verify the impact of different types of social initiatives on distribution marketplaces.

• Computers and Concrete
• /
• 제27권1호
• /
• pp.1-12
• /
• 2021

This work presents a safety assessment of an underground tunnel subjected to a ballistic missile attack employing the numerical approach. For the impact simulation, a box shaped reinforced concrete (RC) structure with a cross section dimension of 8.0×10.0 m under a soil layer that was attacked by a SCUD missile was modeled using finite element (FE) software LS-DYNA. SCUD missile is one of a series of tactical ballistic missiles developed by Soviet Union during the Cold War, which is adopted for a short-range ballistic missile. The developed FE simulation for the penetration depth of the missile impacting into the soil structure was verified from the well-known formula of the penetration prediction. The soil-structure interaction, the soil type, and the impact missile velocity effects on the penetration depth of the missile into the different soil types were investigated. The safety assessment of the underground tunnel was performed with regard to the different depths of the underground tunnel. For each missile velocity and soil type, a specific depth called the unsafe depth was obtained from the analysis results. The structure beneath the soil beyond this depth remains safe. The unsafe depth was found to be increased with the increasing missile velocity.

• International Journal of Computer Science & Network Security
• /
• 제21권4호
• /
• pp.123-130
• /
• 2021

A novel approach for the detection of cheating during e-Exams is presented here using convolutional neural networks (CNN) based systems. This system will help the proctors to identify any kind of uncertain event at the time of online exams, for which most of the government's across the globe are recommending due to the Covid-19 pandemic. Most of the institutions and students across the globe are badly affected by their academic programs and it is a challenging task for universities to conduct examinations using the traditional methods. Therefore, the students are attending most of their classes using different types of third party applications that are available online. However, to conduct online exams the universities cannot rely on these service providers for a long time. Therefore, in this work, a complete setup of the software tools is provided for the students, which can be used by students at their respective laptops/personal computers with strict guidelines from the university. The proposed approach helps most of the universities in Saudi Arabia to maintain their database of different events/activities of students at the time of E-Exams. This method proved to be more accurate and CNN based detection proved to be more sensitive with an accuracy of 97% to detect any kind of uncertain activity of the students at the time of e-Exam.

• The Journal of Asian Finance, Economics and Business
• /
• 제8권2호
• /
• pp.1135-1146
• /
• 2021

The paper aims to examine the relationship between various types of motivation and satisfaction as well as turnover intentions of public university lecturers. Intrinsic motivation is found to affect both satisfaction and turnover, while for extrinsic motivation, only one type has a significant effect on satisfaction. The web-based questionnaire with convenience sampling was chosen to approach target respondents. Data was collected by sending a link of the questionnaire to the lecturers' email. There were 700 questionnaires collected. Of which 681 valid responses, after screening out unusual questionnaires. In order to fulfill these aims, the authors employed structural equation modeling with maximum likelihood method supported by AMOS software with 681 valid samples. The evidence confirms the high correlation between intrinsic motivation and satisfaction for university lecturers which has been shown in previous studies. With regard to extrinsic motivation, extrinsic regulation (material) is found to have a positive influence on satisfaction, meaning financial rewards and job security will lead to his contentment while introjected regulation is a source of dissatisfaction. The results also demonstrate that intrinsic motivation is negatively correlated with turnover intentions. A person who is interested in the job itself is less likely to leave his or her job due to a keen passion for it.

• Structural Engineering and Mechanics
• /
• 제77권4호
• /
• pp.481-493
• /
• 2021

Low rise masonry structures are relatively inexpensive and easier to construct compared to other types of structures such as steel and reinforced concrete buildings. However, masonry structures are relatively heavier and less ductile and more vulnerable to damages in earthquakes. In this research, a new innovative low-cost seismic isolator using steel rings (SISR) is employed to reduce the seismic vulnerability of masonry structures. FEA of a masonry structure, made of concrete blocks is used to evaluate the effect of the proposed SISR on the seismic response of the structure. Two systems, fixed base and isolated from the base with the proposed SISRs, are considered. Micro-element approach and ABAQUS software are used for structural modeling. The nonlinear structural parameters of the SISRs, extracted from a recent experimental study by the authors, are used in numerical modeling. The masonry structure is studied in two separate modes, fixed base and isolated base with the proposed SISRs, under Erzincan and Imperial Valley-06 earthquakes. The accelerated response at the roof level, as well as the deformation in the masonry walls, are the parameters to assess the effect of the proposed SISRs. The results show a highly improved performance of the masonry structure with the SISRs.

• 문화기술의 융합
• /
• 제7권3호
• /
• pp.481-486
• /
• 2021

기계학습 알고리즘은 기준 함수를 채택하여 데이터를 처리하고 학습 모델을 유도한다. 군집분석에서 사용하는 기준 함수는 어떤 형태로든지 선호성을 내포하게 되고 이를 통해 유사한 데이터끼리 묶어 준 후 이를 구성하는 변수와 값들을 특정하여 군집을 정의하게 된다. 군집분석에서 사용하는 카테고리 유용도와 분할 유용도 점수가 군집분석 결과물에 어떤 영향을 주는지를 파악하고 이들이 결과에 어떤 편향성으로 이어지는지를 분석한다. 본 연구는 군집분석에 사용되는 기준 함수의 특성에 따라 결과에 미치는 영향을 파악하기 위해 여러 데이터 세트를 이용해 실험하고 결과를 평가한다.

• Journal of Multimedia Information System
• /
• 제9권1호
• /
• pp.11-20
• /
• 2022

In this paper, we have recently created self-driving cars and self-parking systems in human-friendly cars that can provide high safety and high convenience functions by recognizing the internal and external situations of automobiles in real time by incorporating next-generation electronics, information communication, and function control technologies. And with the development of connected cars, the ITS (Intelligent Transportation Systems) market is expected to grow rapidly. Intelligent Transportation System (ITS) is an intelligent transportation system that incorporates technologies such as electronics, information, communication, and control into the transportation system, and aims to implement a next-generation transportation system suitable for the information society. By combining the technologies of connected cars and Internet of Things with software features and operating systems, future cars will serve as a service platform to connect the surrounding infrastructure on their own. This study creates a research methodology based on the Enhanced Security Model in Self-Driving Cars model. As for the types of attacks, Availability Attack, Man in the Middle Attack, Imperial Password Use, and Use Inclusive Access Control attack defense methodology are used. Along with the commercialization of 5G, various service models using advanced technologies such as autonomous vehicles, traffic information sharing systems using IoT, and AI-based mobility services are also appearing, and the growth of smart transportation is accelerating. Therefore, research was conducted to defend against hacking based on vulnerabilities of smart cars based on artificial intelligence blockchain.

• Journal of Multimedia Information System
• /
• 제9권1호
• /
• pp.33-42
• /
• 2022

In this paper, we have recently created self-driving cars and self-parking systems in human-friendly cars that can provide high safety and high convenience functions by recognizing the internal and external situations of automobiles in real time by incorporating next-generation electronics, information communication, and function control technologies. And with the development of connected cars, the ITS (Intelligent Transportation Systems) market is expected to grow rapidly. Intelligent Transportation System (ITS) is an intelligent transportation system that incorporates technologies such as electronics, information, communication, and control into the transportation system, and aims to implement a next-generation transportation system suitable for the information society. By combining the technologies of connected cars and Internet of Things with software features and operating systems, future cars will serve as a service platform to connect the surrounding infrastructure on their own. This study creates a research methodology based on the Enhanced Security Model in Self-Driving Cars model. As for the types of attacks, Availability Attack, Man in the Middle Attack, Imperial Password Use, and Use Inclusive Access Control attack defense methodology are used. Along with the commercialization of 5G, various service models using advanced technologies such as autonomous vehicles, traffic information sharing systems using IoT, and AI-based mobility services are also appearing, and the growth of smart transportation is accelerating. Therefore, research was conducted to defend against hacking based on vulnerabilities of smart cars based on artificial intelligence blockchain.