• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.1
• /
• pp.70-75
• /
• 2016

Nowadays, there are many research for recognizing face of people using the machine vision. the machine vision is classification and analysis technology using machine that has sight such as human eyes. In this paper, we propose algorithm for classifying human face using this machine vision system. This algorithm consist of Convolutional Neural Network and cascade face detector. And using this algorithm, we classified the face of subjects. For training the face classification algorithm, 2,000, 3,000, and 4,000 images of each subject are used. Training iteration of Convolutional Neural Network had 10 and 20. Then we classified the images. In this paper, about 6,000 images was classified for effectiveness. And we implement the system that can classify the face of subjects in realtime using USB camera.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.4
• /
• pp.1904-1926
• /
• 2016

Iris recognition for biometric personnel identification has gained much interest owing to the increasing concern with security today. The image quality plays a major role in the performance of iris recognition systems. When capturing an iris image under uncontrolled conditions and dealing with non-cooperative people, the chance of getting non-ideal images is very high owing to poor focus, off-angle, noise, motion blur, occlusion of eyelashes and eyelids, and wearing glasses. In order to improve the accuracy of iris recognition while dealing with non-ideal iris images, we propose a novel algorithm that improves the quality of degraded iris images. First, the iris image is localized properly to obtain accurate iris boundary detection, and then the iris image is normalized to obtain a fixed size. Second, the valid region (iris region) is extracted from the segmented iris image to obtain only the iris region. Third, to get a well-distributed texture image, bilinear interpolation is used on the segmented valid iris gray image. Using contrast-limited adaptive histogram equalization (CLAHE) enhances the low contrast of the resulting interpolated image. The results of CLAHE are further improved by stretching the maximum and minimum values to 0-255 by using histogram-stretching technique. The gray texture information is extracted by 1D Gabor filters while the Hamming distance technique is chosen as a metric for recognition. The NICE-II training dataset taken from UBRIS.v2 was used for the experiment. Results of the proposed method outperformed other methods in terms of equal error rate (EER).

• Safety and Health at Work
• /
• v.3 no.4
• /
• pp.287-293
• /
• 2012

Objectives: Metabolic syndrome has received attention as a risk factor for cardiovascular disease, with particular importance attached to visceral fat accumulation, which is associated with lifestyle-related diseases and is strongly correlated with waist circumference. In this study, our aim is to propose waist circumference cut-off values that can be used as a marker for fatty liver based on a sample of workers receiving health checkups in Japan. Methods: This study was conducted in a total of 21,866 workers who underwent periodic health checkups between January 2007 and December 2007. The mean age of the subjects was 47.4 years for men (standard deviation [SD]: 8.0) and 44.7 years for women (SD: 6.9). Evaluation included abdominal ultrasound and measurement of waist circumference, body mass index, fasting blood glucose, triglycerides, high-density lipoprotein cholesterol, and blood pressure. Results: Based on receiver operating characteristic curve analysis, the optimal waist circumference cut-off values were shown as 85.0 cm (sensitivity 0.72, specificity 0.69) for men and 80.0 cm (sensitivity 0.75, specificity 0.78) for women. Conclusion: Abdominal ultrasound is the most efficient means of diagnosing fatty liver, but this examination seldom occurs because the test is not routinely performed at workers' health checkups. In people found to have a high risk of fatty liver, recommendations can be made for abdominal ultrasound based on the waist circumference cut-off values obtained in this study. That is, waist circumference can be used in high risk individuals as an effective marker for early detection of fatty liver.

• Fire Science and Engineering
• /
• v.17 no.4
• /
• pp.86-91
• /
• 2003

Since the air stream in the room controlled by HVAC system affects on he expected response of conventional detectors which are designed in accordance with normal characteristics of air stream in the fire incident, unexpected operation time delay may occur in fire. In order to solve this problem and to improve sensitivity so that to initiate fire in its early stages for minimizing damage and protecting people, we studied and developed Air Sampling Smoke Detector. The Air Sampling Smoke Detector is a kind of active-type fire detection system. it draws air continuously from the protected area through an air sampling pipe network to the smoke density analyzer. This study presents smoke density analysing technique and air intake balancing technique through an air sampling pipe network. As a result of evaluating, Air Sampling Smoke Detector was much more sensitive than conventional smoke detectors that passively wait for smoke to reach them and was not affected by ambient airflow in the room by means of balanced air intake through the sampling holes.

• Fire Science and Engineering
• /
• v.23 no.5
• /
• pp.125-132
• /
• 2009

In this study we could contribute to the development of a precise checking tool with which the reliability of the automatic fire detection systems was enhanced and the safety of the people was ensured. In the same way as the domestic technical standard, the portable smoke sensitivity tester, which was developed in this work, could be used to check the capability of the smoke detectors installed in the field. Its heater inside was warmed up to $400^{\circ}C$ in 40 seconds and the paper as a smoke source was burned to produce smoke for the test. With the Photoelectric smoke detector it was possible to measure and control the smoke concentration in a range from 0%/m to 25%/m. With the adjustment of rpm of the fan, it was possible to keep a constant wind velocity in a range from 20cm/sec to 40cm/sec.

• Journal of KIISE:Computing Practices and Letters
• /
• v.12 no.1
• /
• pp.56-65
• /
• 2006

Recently the study on service robots has been proliferated in many fields, and there are active developments for indoor services such as supporting for elderly people. It is important for robot to recognize objects and situations appropriately for effective and accurate service. Conventional object recognition methods have been based on the pre-defined geometric models, but they have limitations in indoor environments with uncertain situation such as the target objects are occluded by other ones. In this paper we propose a Bayesian network model to reason the probability of target objects for effective detection. We model the relationships between objects by activities, which are applied to non-static environments more flexibly. Overall structure is constructed by combining common-cause structures which are the units making relationship between objects, and it makes design process more efficient. We test the performance of two Bayesian networks for verifying the proposed Bayesian network model through experiments, resulting in accuracy of $86.5\%$ and $89.6\%$ respectively.

• Journal of Korea Water Resources Association
• /
• v.49 no.9
• /
• pp.749-759
• /
• 2016

The aim of this study is to apply and to evaluate the radar-based risk prediction algorithm for damage reduction by sudden localized heavy rain in urban areas. The algorithm is combined with three processes such as "detection of cumulonimbus convective cells that can cause a sudden downpour", "automatic tracking of the detected convective cells", and "risk prediction by considering the possibility of sudden downpour". This algorithm was applied to rain events that people were marooned in small urban stream. As the results, the convective cells were detected through this algorithm in advance and it showed that it is possible to determine the risk of the phenomenon of developing into local heavy rain. When use this risk predicted results for flood prevention operation, it is able to secure the evacuation time in small streams and be able to reduce the casualties.

• The KIPS Transactions:PartA
• /
• v.16A no.4
• /
• pp.255-262
• /
• 2009

We present an autonomous entertainment dolphin robot system based on ubiquitous sensor networks(USN). Generally, It is impossible to apply to USN and GPS in underwater bio-mimetic robots. But An Entertainment dolphin robot which presented in this paper operates on the water not underwater. Navigation of the underwater robot in a given area is based on GPS data and the acquired position information from deployed USN motes with emphasis on user interaction. Body structures, sensors and actuators, governing microcontroller boards, and swimming and interaction features are described for a typical entertainment dolphin robot. Actions of mouth-opening, tail splash or water blow through a spout hole are typical responses of interaction when touch sensors on the body detect users' demand. Dolphin robots should turn towards people who demand to interact with them, while swimming autonomously. The functions that are relevant to human-robot interaction as well as robot movement such as path control, obstacle detection and avoidance are managed by microcontrollers on the robot for autonomy. Distance errors are calibrated periodically by the known position data of the deployed USN motes.

• Korean Journal of Community Nutrition
• /
• v.11 no.6
• /
• pp.779-784
• /
• 2006

Several studies about hospital malnutrition have been reported that about more than 40% of hospitalized patients are having nutritional risk factors and hospital malnutrition presents a high prevalence. People in a more severe nutritional status ended up with a longer length of hospital stay and higher hospital cost. Nutrition screening tools identify individuals who are malnourished or at risk of becoming malnourished and who may benefit from nutritional support. For the early detection and treatment of malnourished hospital patients , few valid screening instruments fur Koreans exist. Therefore, the aim of this study was to develop a simple, reliable and valid malnutrition screening tool that could be used at hospital admission to identify adult patients at risk of malnutrition using medical electrical record data. Two hundred and one patients of the university affiliated medical center were assessed on nutritional status and classified as well nourished, moderately or severely malnourished by a Patient-Generated subjective global assessment (PG-SGA) being chosen as the 'gold standard' for defining malnutrition. The combination of nutrition screening questions with the highest sensitivity and specificity at prediction PG-SGA was termed the nutrition screening index (NSI). Odd ratio, and binary logistic regression were used to predict the best nutritional status predictors. Based on regression coefficient score, albumin less than 3.5 g/dl, body mass index (BMI) less than $18.5kg/m^2$, total lymphocyte count less than 900 and age over 65 were determined as the best set of NSI. By using best nutritional predictors receiver operating characteristic curve with the area under the curve, sensitivity and 1-specificity were analyzed to determine the best optimal cut-off point to decide normal or abnormal in nutritional status. Therefore simple and beneficial NSI was developed for identifying patients with severe malnutrition. Using NSI, nutritional information of the severe malnutrition patient should be shared with physicians and they should be cared for by clinical dietitians to improve their nutritional status.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.10
• /
• pp.1347-1354
• /
• 2018

A bedridden patients, elderly people, and dementia who are subject to special care at a medical institution can not handle the feces themselves and need the help of a guardian or care-giver. In particular, toxic substances are contained in the stools, which can cause eczema, dermatitis and urticaria, so it is important to replace diapers. In this paper, we propose a wearable device design for the detection of excretions in consideration of the various excretion requirements of the elderly. The device is a form in which a module are attached to an adult diaper used in a nursing hospital to detect excreta, and the presence or absence of a wearer can be confirmed by an LED. The measured data is transmitted to the smartphone app in real time via Bluetooth in the module and can be checked for popup notification. The validity of this study was verified by comparing the actual excretion with the data collected through the designed module.