• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.3
• /
• pp.150-159
• /
• 2022

This paper presents an implementation of detecting whether a helmet is worn and there is a fall accident through individual image analysis in real-time from extracting the image objects of several workers active in the industrial field. In order to detect image objects of workers, YOLO, a deep learning-based computer vision model, was used, and for whether a helmet is worn or not, the extracted images with 5,000 different helmet learning data images were applied. For whether a fall accident occurred, the position of the head was checked using the Pose real-time body tracking algorithm of Mediapipe, and the movement speed was calculated to determine whether the person fell. In addition, to give reliability to the result of a falling accident, a method to infer the posture of an object by obtaining the size of YOLO's bounding box was proposed and implemented. Finally, Telegram API Bot and Firebase DB server were implemented for notification service to administrators.

• Journal of Power System Engineering
• /
• v.21 no.5
• /
• pp.41-47
• /
• 2017

To implement Argentina tango dancer-like walking of the humanoid robot, a new trajectory generation scheme based on particle swarm optimization of the blending polynomial is presented. Firstly, the characteristics of Argentina tango walking are derived from observation of tango dance. Secondly, these are reflected in walking pose conditions and cost functions of particle swarm optimization to determine the coefficients of blending polynomial. For the stability of biped walking, zero moment point and reference trajectory of swing foot are also included in cost function. Thirdly, after tango walking cycle is divided into 3 stages with 2 postures, optimal trajectories of ankles, knees and hip of lower body, which include 6 sagittal and 4 coronal angles, are derived in consequence of optimization. Finally, the feasibility of the proposed scheme is validated by simulating biped walking of humanoid robot with derived trajectories under the 3D Simscape environment.

• Journal of Environmental Health Sciences
• /
• v.31 no.3
• /
• pp.192-198
• /
• 2005

The occurrence of toxic cyanobacterial blooms has been reported worldwide and poses a threat to human health through drinking water exposure. The toxins they produce are highly water soluble and can leach into the water body. To eliminate any risk of drinking water exposure, removal of these toxins is essential before the water is consumed. Conventional water treatment techniques such as chlorination, if managed well, can be effectively used to remove some of these toxins, however, saxitoxin and its derivatives pose a problem. Little toxicological data are available to evaluate the real threat of these toxins.

• Korean Journal of Veterinary Service
• /
• v.28 no.1
• /
• pp.23-30
• /
• 2005

Ethylene glycol (EG) intoxication in dogs can pose a diagnostic challenge. We investigated the effect of a lethal dose of EG to the Shepherd on blood, urine test, and histopathologic observation. Shepherds were treated orally with EG at $10\;m{\ell}/kg$ body weight. Clinically, EG caused the vomition, depression of central brain. Anion gap was increased, which was decreased pH of blood, The blood osmorality was increased at 3 hours, and BUN and creatinine was increased at 24 hour significantly. The pH of urine decreased. The crystal of calcium oxalate appeared at 6 hours. The amount of crystals increased at 12 and 24 hours. Microscopically, the degeneration of proximal tubules were shown at 1 hour and finally progressed as acute tubular nephrosis at 24 hours. These results suggest that the blood and urine test accompanied with histopathological examination be helpful to investigate the EG intoxication.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.5
• /
• pp.611-615
• /
• 2010

In this paper, we propose a method that exaggerate a character motion using quadratic deformation. While the previous methods tend to exaggerate a rotational motion of an individual joint angle, our method attempt to model the poses of a whole body at each frame and exaggerate those, so that the whole-pose action of the character can be exaggerated. Our method can be computed in real-time, and prevents a joint motion that rotates unexpected direction.

• The Journal of Korea Robotics Society
• /
• v.4 no.4
• /
• pp.265-272
• /
• 2009

IVarious driving mechanisms to adapt to uneven environment have been developed for many urban search and rescue (USAR) missions. A tracked mechanism has been widely used to maintain the stability of robot's pose and to produce large traction force on uneven terrain in this research area. However, it has a drawback of low energy efficiency due to friction force when rotating. Moreover, single tracked mechanism can be in trouble when the body gets caught with high projections, so the track doesn't contact on the ground. A transformable tracked mechanism is proposed to solve these problems. The mechanism is designed with several articulations surrounded by tracks, used to generate an attack angle when the robot comes near obstacles. The stair climbing ability of proposed robot was analyzed since stairs are one of the most difficult obstacles in USAR mission. Stair climbing process is divided into four separate static analysis phases. Design parameters are optimized according to geometric limitations from the static analysis. The proposed mechanism was produced from optimized design parameters, and demonstrated in artificially constructed uneven environment and the actual stairway.

• Journal of Multimedia Information System
• /
• v.4 no.1
• /
• pp.43-50
• /
• 2017

In this paper, we propose a method which detects the nose and face of certain human by using the depth image. The proposed method has advantages of the low computational complexity and the high accuracy even in dark environment. Also, the detection accuracy of nose and face does not change in various postures. The proposed method first locates the locally protruding part from the depth image of the human body captured through the depth camera, and then confirms the nose through the depth characteristic of the nose and surrounding pixels. After finding the correct pixel of the nose, we determine the region of interest centered on the nose. In this case, the size of the region of interest is variable depending on the depth value of the nose. Then, face region can be found by performing binarization using the depth histogram in the region of interest. The proposed method can detect the nose and the face accurately regardless of the pose or the illumination of the captured area.

• International Journal of Costume and Fashion
• /
• v.16 no.2
• /
• pp.1-24
• /
• 2016

This study investigated the ideal female beauty in advertisement images through the iconological comprehension of Renaissance portraits and explored the longitudinal change of the ideal female beauty in modern Korean society. A total of 146 cover images on Hyang-Jang magazine, which is the company magazine of AMORE PACIFIC Group, from 1972 to 2012 were selected as the data. These images were divided into 10-year units for analysis using iconological criteria, which were pose, shape of eyes and lip, and hairstyle. The way presenting the cover images on Hyang-Jang magazine had changed in order to emphasize female body, self-satisfaction and independence of female, and sexual attractiveness of female. The results of this study implied that the change of ideal female beauty had been affected by socio-cultural contexts of modern Korean society that has industrialized and democratized in a short period of time.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.3 no.4
• /
• pp.7-12
• /
• 2004

Inspection and analysis are essential process to determine whether a completed product is in given specification or not. Analysis of products with very complicated shape is difficult to carry out direct comparison between inspected coordinate and designed coordinates. So process called as matching or registrations is needed to solve this problem. By defining error between two coordinates and minimizing the error, registration is done. Registration consists of translation, rotation and scale transformations. Error must be defined to express feature of inspected product. In this paper, registration algorithm is developed to determine pose of sub-frame at assembly with body of automobile by defining error between two coordinates considering geometric feature of sub-frame.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.892-894
• /
• 2005

본 논문에서는 스테레오 영상에서 깊이 정보를 추출하여 사람의 자세를 학습된 2차원 깊이 영상들의 선형 결함으로 표현하여 3차원 인체 모델을 재구성하는 방법을 제안한다. 한 장의 2차원 깊이 영상으로 최소 제곱법을 이용하여 프로토타입 깊이 영상의 선형 결합으로 표현되는 최적의 계수를 찾을 수 있다. 입력된 깊이 영상의 3차원 인체 모델은 프로토타입 깊이 영상에서 예측된 계수를 적용하여 생성한다. 학습 단계에서는 데이터를 계층적으로 나누어 모델을 생성한다. 또한, 재구성 단계에서는 실루엣 영상과 깊이 영상으로부터 계층적으로 나누어진 학습 데이터를 이용하여 3차원 인체 자세를 재구성한다. 학습 및 재구성의 마지막 단계에서는 실루엣 영상 대신 깊이 영상을 이용하여 3차원 인체 모델을 재구성한다. 한 장의 실루엣 영상을 이용하면 영상의 노이즈에 민감하기 때문에 재구성 단계의 상위 레벨에서는 실루엣 영상의 누적 영상을 이용한다. 실험 결과는 제안된 방법이 효율적으로 3차원 인체 자세를 재구성함을 보여준다.