• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.5
• /
• pp.977-984
• /
• 2022

Recently, with the economic development of Southeast Asia, the use of information devices is widely spreading, and the demand for application services using intelligent character recognition is increasing. This paper discusses deep learning-based feature extraction and recognition of Myanmar, one of the Southeast Asian countries. Myanmar alphabet (33 letters) and Myanmar numerals (10 numbers) are used for feature extraction. In this paper, the number of nine features are extracted and more than three new features are proposed. Extracted features of each characters and numbers are expressed with successful results. In the recognition part, convolutional neural networks are used to assess its execution on character distinction. Its algorithm is implemented on captured image data-sets and its implementation is evaluated. The precision of models on the input data set is 96 % and uses a real-time input image.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.17 no.5
• /
• pp.330-338
• /
• 2024

With the growing interest in rehabilitation therapy and exercise programs, there is an increasing need for smart content that simultaneously addresses both health and engagement. Particularly, exercises performed in a state of physical imbalance carry a high risk of injury, making it essential to detect and integrate balance into the training process. This paper proposes Rehabilitation Training program that combines a pressure platform with virtual reality (VR) technology to address this issue. The program enables users to perform exercises such as squats, stationary walking, and forward-backward walking in a VR environment, utilizing real-time foot pressure data captured through a pressure mat. Additionally, an algorithm based on YOLOv8-pose extracted skeletal coordinates is proposed to assess body balance and automatically count squat repetitions. The experimental results showed an average accuracy of 87.9% for each posture, confirming that users can be provided with a safer, more efficient, and immersive training experience through this approach.

• Journal of the Korea Computer Graphics Society
• /
• v.23 no.3
• /
• pp.1-8
• /
• 2017

In this paper, we introduce a technique to retarget human motion data to the humanoid body in a constrained environment. We assume that the given motion data includes detailed interactions such as holding the object by hand or avoiding obstacles. In addition, we assume that the humanoid joint structure is different from the human joint structure, and the shape of the surrounding environment is different from that at the time of the original motion. Under such a condition, it is also difficult to preserve the context of the interaction shown in the original motion data, if the retargeting technique that considers only the change of the body shape. Our approach is to separate the problem into two smaller problems and solve them independently. One is to retarget motion data to a new skeleton, and the other is to preserve the context of interactions. We first retarget the given human motion data to the target humanoid body ignoring the interaction with the environment. Then, we precisely deform the shape of the environmental model to match with the humanoid motion so that the original interaction is reproduced. Finally, we set spatial constraints between the humanoid body and the environmental model, and restore the environmental model to the original shape. To demonstrate the usefulness of our method, we conducted an experiment by using the Boston Dynamic's Atlas robot. We expected that out method can help the humanoid motion tracking problem in the future.

• Journal of the Korea Computer Graphics Society
• /
• v.30 no.3
• /
• pp.189-198
• /
• 2024

Research combining deep learning-based models and physical simulations is making important advances in the medical field. This extracts the necessary information from medical image data and enables fast and accurate prediction of deformation of the skeleton and soft tissue based on physical laws. This study proposes a system that integrates Neural Radiance Fields (NeRF), Position-Based Dynamics (PBD), and Parallel Resampling to generate 3D volume data, and deform and visualize them in real-time. NeRF uses 2D images and camera coordinates to produce high-resolution 3D volume data, while PBD enables real-time deformation and interaction through physics-based simulation. Parallel Resampling improves rendering efficiency by dividing the volume into tetrahedral meshes and utilizing GPU parallel processing. This system renders the deformed volume data using ray casting, leveraging GPU parallel processing for fast real-time visualization. Experimental results show that this system can generate and deform 3D data without expensive equipment, demonstrating potential applications in engineering, education, and medicine.

• The Korean Fashion and Textile Research Journal
• /
• v.14 no.5
• /
• pp.834-845
• /
• 2012

This study was conducted to provide basic information on developing highschool-uniform pants with more suitable fit and to suggest a sizing system for highschool girls' uniform pants that adequately reflects their body figures. To understand the features of high school girls' lower body type, the body measurement values of 833 girls from 16~18 years of age based on Size Korea(2010) were analyzed statistically. For the classification of lower body type for high school girls, a factor analysis and cluster analysis were conducted. The collected data were processed with the programs SPSS 18.0 for windows. The results in this study are follows: The lower body types for high school girls were divided into 3 groups. Body Type A is average stature but the biggest circumference, Type B is the biggest stature and the medium body type, Type C is the smallest stature and skeleton structure. KS size intervals were used for frequency distribution of height and waist for the lower body. Sizing system of the uniform company and frequency distribution of sizes were compared. Using the two-way distribution of highschool girls' waist circumference and hip circumference, sizing system considering body type distribution and high frequency distribution section of sizes was suggested. This study established new sizing system depending on lower body fixed as 26 number of sizes. The most suitable standard is fixed as 12 number of sizes ; 64-88, 64-91, 67-88, 67-91, 67-94, 70-91, 70-94, 70-97, 73-94, 73-97, 76-97, 76-100. The coverage is also calculated. And the coverage of new standard was 63.5%. The continuous study on the uniform pants sizing system of the obesity types is required.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.11
• /
• pp.801-808
• /
• 2013

In the virtual space with the interactive 3D contents, the degree of mental satisfaction is determined by how fully it reflect the real world. There are a few factors for getting the high completeness of virtual space. The first is the modeling technique with high-polygons and high-resolution textures which can heighten an visual effect. The second is the functionality. It is about how realistic represents dynamic actions between the virtual space and the user or the system. Although the studies on the techniques for animating and controlling the virtual characters have been continued, there are problems such that the long production time, the high cost, and the animation without expected behaviors. This paper suggest a method of behavior control of animation by designing the optimized skeleton which produces the movement of character and applying the procedural technique using physical law and mathematical analysis. The proposed method is free from the constraint on one-to-one correspondence rules, and reduce the production time by controlling the simple parameters, and to increase the degree of visual satisfaction.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.3
• /
• pp.558-567
• /
• 2017

In this research, we developed an efficient real-time human exercise pose estimation system using multiple Kinects. The main objective of this system is to measure and recognize the user's posture (such as knee curl or lunge) more accurately by employing Kinects on the front and the sides. Especially it is designed as an extensible and modular method which enables to support various additional postures in the future. This system is configured as multiple clients and the Unity3D server. The client processes Kinect skeleton data and send to the server. The server performs the multiple-Kinect calibration process and then applies the pose estimation algorithm based on the Kinect-based posture recognition model using feature extractions and the weighted averaging of feature values for different Kinects. This paper presents the design and implementation of the human exercise pose estimation system using multiple Kinects and also describes how to build and execute an interactive Unity3D exergame.

• Journal of Nutrition and Health
• /
• v.32 no.1
• /
• pp.50-63
• /
• 1999

This study was conducted with 20 female gymnasts and 23 age-matched controls to examine the relationship of diet, menstrual function and bone mineral density (BMD). The results obtained are summarized as follows : Energy intake of gymnasts was 968.9$\pm$421.4kcal, and energy expenditure was 2091.4$\pm$361kcal showing negative energy balance(-1,122.5$\pm$534.6kcal). The average intakes of calcium, iron, vitamin A, thiamin, riboflavin and niacin did not meet the Recommended Dietary Allowances for their age groups. Mean age at menarche in gymnasts is 15.8$\pm$1.2 years compared with 11.8$\pm$2.8 years in age-matched controls. The profile of estradiol, progesterone, and luteinizing hormone was lower than age-matched controls but not significant. Athletic amenorrheic gymnasts(n=12) have the menstrual irregularity(n=10) and amenorrhea(n=2). A number of variables as such nutritional deficiency in diet, negative energy blasnce and hypogonadotropic hormonal status were included. The bone mineral density (BMD) of female gymnasts were significantly higher than controls for the lumbar neck(p<0.001), trochanter(p<0.01), and Ward's triangle(p<0.001), but there were no significant differences for the lumbar spine and forearm. The lumbar spine BMD had a positive correlation with age and lean body weight. The femoral neck BMD was significantly associated with age, group and lean body mass. The trochanter BMD had significant relationship with group, body mass index, energy expenditure and follicular stimulating hormone. Ward's triangle BMD were related to body mass index and follicular stimulating hormone. The significant association was deterced between forearm BMD and age and lean body weight. The major finding of this investigation is that the BMD of gymnasts were higher than age-matched controls despite the fact that gymnasts as a group had inadequate dietary calcium and a higher propensity to have an interruption of their menstrual cycle. These data indicate that grymnsts involved in sports producing significant impact loading on the skeleton had greater femoral neck, trochanter and Ward's triangle bone density than age-matched controls.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.41 no.11
• /
• pp.1035-1045
• /
• 2017

Classification of age and gender has been carried out through different approaches such as facial-based and audio-based classifications. One of the limitations of facial-based methods is the reduced recognition rate over large distances, while another is the prerequisite of the faces to be located in front of the camera. Similarly, in audio-based methods, the recognition rate is reduced in a noisy environment. In contrast, gait-based methods are only required that a target person is in the camera. In previous works, the view point of a camera is only available as a side view and gait data sets consist of a standard gait, which is different from an ordinary gait in a real environment. We propose a feature extraction method using skeleton models from an RGB-D sensor by considering characteristics of age and gender using ordinary gait. Experimental results show that the proposed method could efficiently classify age and gender within a target group of individuals in real-life environments.

• Structural Engineering and Mechanics
• /
• v.75 no.5
• /
• pp.571-583
• /
• 2020

Permeation grouting is of great significance for consolidating geo-materials without disturbing the original geo-structure. To dip into the filtration-induced pressure increment that dominates the grout penetration in permeation grouting, nonlinear filtration coefficients embedded in a convection-filtration model were proposed, in which the volume of cement particles in grout and the deposited particles of skeleton were considered. An experiment was designed to determine the filtration coefficients and verify the model. The filtration coefficients deduced from experimental data were used in simulation, and the modelling results matched well with the experimental ones. The pressure drop revealed in experiments and captured in modelling demonstrated that the surge of inflow pressure lagged behind the stoppage of flow channels. In addition, both the consideration of the particles loss in liquid grout and the number of filtrated particles on pore walls presented an ideal trend in filtration rate, in which the filtration rate first rose rapidly and then reached to a steady plateau. Finally, this observed pressure drop was extended to the grouting design which alters the water to cement (W/C) ratio so as to alleviate the filtration effect. This study offers a novel insight into the filtration behaviour and has a practical meaning to extend penetration distance.