• Therapeutic Science for Rehabilitation
• /
• v.12 no.2
• /
• pp.55-68
• /
• 2023

Objective : This study aims to analyze the effects of visual perception intervention by systematically reviewing the studies that applied visual perception intervention to children with cerebral palsy. Methods : The databases used were PubMed, EMbase, Science Direct, ProQuest, Koreanstudies Information Service System (KISS), Research Information Sharing Service (RISS), and the National Assembly Library. The keywords used were cerebral palsy, CP, and visual perception. According to the PRISMA flowchart, 10 studies were selected from among studies published from January 1, 2012 to March 30, 2022. The quality level of the selected studies, the demographic characteristics of study participants, the effectiveness of interventions, area and strategies of intervention, assessment tools to measure the effectiveness of interventions, and risk of bias were analyzed. Results : All selected studies confirmed that visual perception intervention was effective in improving visual perception function. In addition, positive results were shown in upper extremity function, activities of daily living, posture control, goal achievement, and psychosocial areas as well as visual perception function. The eye-hand coordination area was intervened in all studies. Conclusion : In visual perception intervention, It is necessary to evaluate the visual perception function by area, and apply systematically graded customized interventions for each individual.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.12
• /
• pp.489-498
• /
• 2022

Recently, the textile industry, which is labor-intensive, is attempting to reduce process costs and optimize quality through artificial intelligence. However, the fiber spinning process has a high cost for data collection and lacks a systematic data collection and processing system, so the amount of accumulated data is small. In addition, data imbalance occurs by preferentially collecting only data with changes in specific variables according to the purpose of fiber spinning, and there is an error even between samples collected under the same fiber spinning conditions due to difference in the measurement environment of physical properties. If these data characteristics are not taken into account and used for AI models, problems such as overfitting and performance degradation may occur. Therefore, in this paper, we propose an outlier handling technique and data augmentation technique considering the characteristics of the spinning process data. And, by comparing it with the existing outlier handling technique and data augmentation technique, it is shown that the proposed technique is more suitable for spinning process data. In addition, by comparing the original data and the data processed with the proposed method to various models, it is shown that the performance of the tensile tenacity and elongation prediction model is improved in the models using the proposed methods compared to the models not using the proposed methods.

• Journal of Broadcast Engineering
• /
• v.27 no.1
• /
• pp.104-123
• /
• 2022

Style transfer based on neural network provides very high quality results by reflecting the high level structural characteristics of images, and thereby has recently attracted great attention. This paper deals with the problem of resolution limitation due to GPU memory in performing such neural style transfer. We can expect that the gradient operation for style transfer based on partial image, with the aid of the fixed size of receptive field, can produce the same result as the gradient operation using the entire image. Based on this idea, each component of the style transfer loss function is analyzed in this paper to obtain the necessary conditions for partitioning and padding, and to identify, among the information required for gradient calculation, the one that depends on the entire input. By structuring such information for using it as auxiliary constant input for partition-based gradient calculation, this paper develops a recursive algorithm for super high-resolution image style transfer. Since the proposed method performs style transfer by partitioning input image into the size that a GPU can handle, it can perform style transfer without the limit of the input image resolution accompanied by the GPU memory size. With the aid of such super high-resolution support, the proposed method can provide a unique style characteristics of detailed area which can only be appreciated in super high-resolution style transfer.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.6 no.3
• /
• pp.383-391
• /
• 2016

Ballast water provides stability and manoeuvrability to a ship. Foreign harmful aquatic organisms, which were transferred by ballast water, cause disturbing ecosystem. In order to minimize transference of foreign harmful aquatic organisms, IMO(International Maritime Organization) adopted the International Convention for the Control and Management of Ship's Ballast Water and Sediments in 2004. If the convention take effect, a port authority might need to check that ballast water is properly disposed of. In this paper, we propose a method of counting harmful aquatic organisms in ballast water thorough image processing. We extracted three samples from the ballast water that had been collected at Busan port in Korea. Then we made three grey-scale images from each sample as experimental data. We made a comparison between the proposed method and CellProfiler which is a well known cell-counting program based on image processing. Setting of CellProfiler is empirically chosen from the result of cell count by an expert. After finding a proper threshold for each image at which the result is similar to that of CellProfiler, we used the average value as the final threshold. Our experimental results showed that the proposed method is simple but about ten times faster than CellProfiler without loss of the output quality.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.12
• /
• pp.871-879
• /
• 2017

The purpose of this study is to propose a Busan tourism industry in which the 2016 OECD Tourism policy and ICT convergence platform are applied. OECD proposed 3 policies to promote the tourism industry: First, to maintain the competitiveness of the tourism industry as well as improve its efficiency and sustainability, second, to establish a seamless traffic system, and third, to build a response to the sharing economy. Centering on the OECD's three policies, we propose the developmental possibilities of tourism in Busan. At the same time, we suggest the necessity to build an ICT convergence platform that will help foster the industry. In building an ICT convergence platform, we especially focus on the necessity of: 1. Sharing and creating experience-based interactive contents on the software side, and 2. Developing high quality user experience (UX) and providing a data analysis-based customized service on the hardware side. In addition, we insist on the establishment of the Tourism Promotion Agency for the continuous performance and management of Busan tourism industry. The study ultimately suggests that the construction of ICT convergence platform based on OECD tourism policy can result in the expected outcomes of high effects with low cost for both consumers and suppliers related to the tourism industry.

• Korean Journal of Artificial Intelligence
• /
• v.11 no.2
• /
• pp.39-44
• /
• 2023

The creativity of thesis is that the significance of cyber security challenges in blockchain. The variety of enterprises, including those in the medical market, are the targets of cyberattacks. Hospitals and clinics are only two examples of medical facilities that are easy targets for cybercriminals, along with IoT-based medical devices like pacemakers. Cyberattacks in the medical field not only put patients' lives in danger but also have the potential to expose private and sensitive information. Reviewing and looking at the present and historical flaws and vulnerabilities in the blockchain-based IoT and medical institutions' equipment is crucial as they are sensitive, relevant, and of a medical character. This study aims to investigate recent and current weaknesses in medical equipment, of blockchain-based IoT, and institutions. Medical security systems are becoming increasingly crucial in blockchain-based IoT medical devices and digital adoption more broadly. It is gaining importance as a standalone medical device. Currently the use of software in medical market is growing exponentially and many countries have already set guidelines for quality control. The achievements of the thesis are medical equipment of blockchain-based IoT no longer exist in a vacuum, thanks to technical improvements and the emergence of electronic health records (EHRs). Increased EHR use among providers, as well as the demand for integration and connection technologies to improve clinical workflow, patient care solutions, and overall hospital operations, will fuel significant growth in the blockchain-based IoT market for linked medical devices. The need for blockchain technology and IoT-based medical device to enhance their health IT infrastructure and design and development techniques will only get louder in the future. Blockchain technology will be essential in the future of cybersecurity, because blockchain technology can be significantly improved with the cybersecurity adoption of IoT devices, i.e., via remote monitoring, reducing waiting time for emergency rooms, track assets, etc. This paper sheds the light on the benefits of the blockchain-based IoT market.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.3
• /
• pp.825-834
• /
• 2023

The demands of people who purchase fashion products on Internet shopping are gradually increasing, and attempts are being made to provide user-friendly images with 3D contents and web 3D software instead of pictures and videos of products provided. As a reason for this issue, which has emerged as the most important aspect in the fashion web shopping industry, complaints that the product is different when the product is received and the image at the time of purchase has been heightened. As a way to solve this problem, various image processing technologies have been introduced, but there is a limit to the quality of 2D images. In this study, we proposed an automatic conversion technology that converts 2D images into 3D and grafts them to web 3D technology that allows customers to identify products in various locations and reduces the cost and calculation time required for conversion. We developed a system that shoots a mannequin by placing it on a rotating turntable using only 8 cameras. In order to extract only the clothing part from the image taken by this system, markers are removed using U-net, and an algorithm that extracts only the clothing area by identifying the color feature information of the background area and mannequin area is proposed. Using this algorithm, the time taken to extract only the clothes area after taking an image is 2.25 seconds per image, and it takes a total of 144 seconds (2 minutes and 4 seconds) when taking 64 images of one piece of clothing. It can extract 3D objects with very good performance compared to the system.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.10
• /
• pp.461-470
• /
• 2023

While speech animation generation employing deep learning has been actively researched for English, there has been no prior work for Korean. Given the fact, this paper for the very first time employs supervised deep learning to generate Korean speech animation. By doing so, we find out the significant effect of deep learning being able to make speech animation research come down to speech recognition research which is the predominating technique. Also, we study the way to make best use of the effect for Korean speech animation generation. The effect can contribute to efficiently and efficaciously revitalizing the recently inactive Korean speech animation research, by clarifying the top priority research target. This paper performs this process: (i) it chooses blendshape animation technique, (ii) implements the deep-learning model in the master-servant pipeline of the automatic speech recognition (ASR) module and the facial action coding (FAC) module, (iii) makes Korean speech facial motion capture dataset, (iv) prepares two comparison deep learning models (one model adopts the English ASR module, the other model adopts the Korean ASR module, however both models adopt the same basic structure for their FAC modules), and (v) train the FAC modules of both models dependently on their ASR modules. The user study demonstrates that the model which adopts the Korean ASR module and dependently trains its FAC module (getting 4.2/5.0 points) generates decisively much more natural Korean speech animations than the model which adopts the English ASR module and dependently trains its FAC module (getting 2.7/5.0 points). The result confirms the aforementioned effect showing that the quality of the Korean speech animation comes down to the accuracy of Korean ASR.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.6
• /
• pp.85-96
• /
• 2023

In a reality where negative perceptions and social prejudices towards individuals with mental illnesses persist, the absence and lack of social support systems acquired through personal intimate social networks can be considered a significant hindrance to the quality of life for those with mental disorders. Therefore, this study examined the impact of the characteristics of social networks and social support on the life satisfaction of individuals with mental illnesses. A survey was conducted with 180 patients from seven mental health treatment facilities in the Daegu area. The data were analyzed using MANOVA, hierarchical multiple regression analysis, and Sobel test for mediation analysis with SPSS 25 software. The key findings of the study revealed that the intimacy aspect of social networks had a positive and significant effect on social support and life satisfaction. On the other hand, the size of the social network showed a negative influence on social support. Furthermore, social support partially mediated the relationship between the intimacy of social networks and life satisfaction, and it fully mediated the relationship between the size of social networks and life satisfaction. Based on these research outcomes, practical and policy-related recommendations are provided to enhance life satisfaction through increased social support for this population.

• Journal of Industrial Convergence
• /
• v.21 no.9
• /
• pp.11-21
• /
• 2023

There have been many studies to improve the efficiency of the CG production process, but it was not easy to overcome the problem that it was difficult to check the result in the middle of work and it took a lot of time for rendering. However, as the possibility of using Unreal Live Link, which can check the result in real-time, is increasing, expectations for improving the efficiency of the production process are rising. This study analyzed the efficiency of the 3D animation production process using Unreal Live Link. To this end, modeling, rigging, animation, and layout work were done in Maya, and the final output image sequence was rendered in Unreal Engine through Unreal Live Link. And the difference between this production process and the existing production process in which the final output image sequence is rendered in the 3D software itself was compared and analyzed. As a result of the analysis, unlike the traditional 3D animation production process, it was possible to check the final work result in real-time by proceeding with the work through a high-quality viewport screen, and it was found that the efficiency of work was maximized by deriving the final result through real-time screen capture. Recently, the use of game engines in the 3D animation and film industry is gradually increasing, and the efficiency of work is expected to be maximized if Unreal Live Link is used.