• Smart Structures and Systems
• /
• v.28 no.6
• /
• pp.779-789
• /
• 2021

Cracks in railway sleeper are an inevitable condition and has a significant influence on the safety of railway system. Although the technology of railway sleeper condition monitoring using machine learning (ML) models has been widely applied, the crack recognition accuracy is still in need of improvement. In this paper, a two-stage method using edge detection and convolutional neural network (CNN) is proposed to reduce the burden of computing for detecting cracks in railway sleepers with high accuracy. In the first stage, the edge detection is carried out by using the 3×3 neighborhood range algorithm to find out the possible crack areas, and a series of mathematical morphology operations are further used to eliminate the influence of noise targets to the edge detection results. In the second stage, a CNN model is employed to classify the results of edge detection. Through the analysis of abundant images of sleepers with cracks, it is proved that the cracks detected by the neighborhood range algorithm are superior to those detected by Sobel and Canny algorithms, which can be classified by proposed CNN model with high accuracy.

• Journal of muscle and joint health
• /
• v.7 no.2
• /
• pp.241-257
• /
• 2000

Rheumatoid arthritis as one of the chronic illness requiring management in long period of time puts great burden to patients, their family and society. For patients with chronic illnesses, providing a social support is important and the most important source comes from spouses. Therefore we assessed burden of husbands of female rheumatoid arthritic patients and also found out the factors affecting burden. The sample of study was 107 female rheumatoid arthritic patients and their spouses. The tool of assessing spouses' burden was the revised version of subjective and objective parameters developed by Montgomery et al.(1985). The results are as follows: 1. General characteristics of patients and spouses: The mean age of the patients was 48 years. Educational level of patients was high school 41.1%. The mean age of the spouses was 51years. Educational level of spouses was mostly high school(40.2%) and college(29.9%) graduate. The mean marital period was 23.4years. Average income per month was 1,609,000 won. The average duration since diagnosis was 9.4years. As a therapy, 67.3% chose standard drug therapy. Average rating of discomfort by patient was 3.05(range 1-5) and that of severity was 3.48 and that of dependency was 2.58. The husband's rating of their spouses disease severity was 3.68. 2. Husbands' burden: The average burden in subjective items was 21.61(range 6-36) and objective items was 35.24(range 10-60). The average of total burden was 56.59(range 16-96). 3. Husband's total burden correlated with patient's age, educational level of patients, therapy method, patient's level of discomfort, patient's severity, patient's level of dependence, husband's recognition of level of severity in statistical level. Husband's objective burden correlated with patient's age, educational level of patient, patient's level of discomfort, husband's recognition of level of severity. Husband's subjective burden correlated with patient's age, educational level of patients, therapy method, patient's severity, patient's level of dependence, husband's recognition of level of severity. 4. Linear correlation analysis on burden: The husbands' total burden is explained in 37 7% by husband's recognition of level of severity and husband's age. The husbands' objective burden is explained in 31.2% by patient's level of dependence, husband's age, husband's recognition of level of severity. The husbands' subjective burden is explained in 26.7% by husband's recognition of level of severity and patient's age. In conclusion, husbands' level of burden is affected by many factors and therefore nursing strategy for relieving burden of middle aged husbands should be individualized taking these factors into consideration.

• The Journal of Korea Robotics Society
• /
• v.4 no.4
• /
• pp.257-264
• /
• 2009

This paper proposes a novel face detection method that finds tiny faces located at a long range even with low-resolution input images captured by a mobile robot. The proposed approach can locate extremely small-sized face regions of $12{\times}12$ pixels. We solve a tiny face detection problem by organizing a system that consists of multiple detectors including a mean-shift color tracker, short- and long-rage face detectors, and an omega shape detector. The proposed method adopts the long-range face detector that is well trained enough to detect tiny faces at a long range, and limiting its operation to only within a search region that is automatically determined by the mean-shift color tracker and the omega shape detector. By focusing on limiting the face search region as much as possible, the proposed method can accurately detect tiny faces at a long distance even with a low-resolution image, and decrease false positives sharply. According to the experimental results on realistic databases, the performance of the proposed approach is at a sufficiently practical level for various robot applications such as face recognition of non-cooperative users, human-following, and gesture recognition for long-range interaction.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.8
• /
• pp.1813-1819
• /
• 2011

In this research, the location positioning system supposed is concerned with range recognition technology using phase and magnitude of radio wave and adding technology of image histogram by vision. By the proposed technology, we design the radio transmitter and receiver and realize the measurement system, and save the data in disk that is earned from 900Mhz RF signal, middle frequency 450Khz of analog signal. Range information is earned the data through digital signal processing of IF signal. For the estimation of range measured, we analyze the difference between real range and measurement range, and also suggest the method to improve the measurement error using average processing and amplitude properties.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.10a
• /
• pp.699-701
• /
• 2017

Typical operating method for long-range observation cameras are to detect the target at a wide angle of view and to recognize/identify the target with a telephoto angle of view. And the detection/recognition range performance is an important item to evaluate the performance of the defense infrared camera. To increased the detection range performance, the camera's field of view should be narrowed. Due to the narrow field of view, the probability of finding target is relatively low. In this paper, we propose a method to search for target by providing a wide angle view while maintaining detection range performance. M&S and optimized design were used to develop infrared camera with extended field of view and the results of the test summarized.

• Journal of Korea Multimedia Society
• /
• v.21 no.5
• /
• pp.626-637
• /
• 2018

Human exchanges information not only through words, but also through body gesture or hand gesture. And they can be used to build effective interfaces in mobile, virtual reality, and augmented reality. The past 2D gesture recognition research had information loss caused by projecting 3D information in 2D. Since the recognition of the gesture in 3D is higher than 2D space in terms of recognition range, the complexity of gesture recognition increases. In this paper, we proposed a real-time gesture recognition deep learning model and application in 3D space using deep learning technique. First, in order to recognize the gesture in the 3D space, the data collection is performed using the unity game engine to construct and acquire data. Second, input vector normalization for learning 3D gesture recognition model is processed based on deep learning. Thirdly, the SELU(Scaled Exponential Linear Unit) function is applied to the neural network's active function for faster learning and better recognition performance. The proposed system is expected to be applicable to various fields such as rehabilitation cares, game applications, and virtual reality.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.05a
• /
• pp.74-77
• /
• 2015

WiSee is by utilizing the frequency of Wi-Fi(802.11n/ac), a technique for performing the operation recognized by the user's gesture. Current motion recognition scheme are using a dedicated device (leaf motion, Kinekuto) and the recognition range is 30cm ~ 3.5m. also For recognition range increases the narrow recognition rate, there is inconvenience for maintaining a limited distance. But WiSee is used by Wi-Fi it is possible to anywhere motion recognition if available location. Permeability also has advantages as compared with the conventional recognition method. In this paper I take a look at the operation process and the recent trend of WiSee.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.40 no.6
• /
• pp.31-40
• /
• 2003

Most face recognition systems are based on 2D images and applied in many applications. However, it is difficult to recognize a face when the pose varies severely. Therefore, head pose estimation is an inevitable procedure to improve recognition rate when a face is not frontal. In this paper, we propose a novel head pose estimation algorithm for 3D face recognition. Given the 3D range image of an unknown face as an input, we automatically extract facial feature points based on the face curvature. We propose an Error Compensated Singular Value Decomposition (EC-SVD) method based on the extracted facial feature points. We obtain the initial rotation angle based on the SVD method, and perform a refinement procedure to compensate for remained errors. The proposed algorithm is performed by exploiting the extracted facial features in the normaized 3D face space. In addition, we propose a 3D nearest neighbor classifier in order to select face candidates for 3D face recognition. From simulation results, we proved the efficiency and validity of the proposed algorithm.

• The Korean Journal of Health Service Management
• /
• v.5 no.3
• /
• pp.13-23
• /
• 2011

This paper investigated the operating environment for the representative of each agency and the facility workers on the basis of analytical result of recognition changes of the operating environment changes under the operating the long-term care insurance. It was described plans to take positive effect on the operating as follows. The first, on the result of regression analysis, the service administrative range takes the biggest effect on the general recognition of executing the long-term care insurance off and on. The affirmative recognition of the service administrative range had the general recognition on the system be positive effect. But the operator of facility asserts that the care manager's professionalism related quality of service be strengthened. The second, on the result of regression analysis, in the financial accounting administrative it is revealed the more positive recognition it is, the more positive effects it has. From the difference verification of an operation size from operation subject, the small operation size and personal facility recognize the long term care insurance positively. On the other side the facilities where the operation size is big recognize the system negatively. The long-term care facility should rearrange a support program newly and the government needs to promote the donation activity, because it is needed to reduce the financial burden of facilities.

• Journal of the Korea Convergence Society
• /
• v.11 no.6
• /
• pp.1-6
• /
• 2020

In this study, research on character recognition, which is one of the fields of computer vision, was conducted. Optical character recognition, which is one of the most widely used character recognition techniques, suffers from decreasing recognition rate if the recognition target deviates from a certain standard and format. Hence, this study aimed to address this limitation by applying deep learning techniques to character recognition. In addition, as most character recognition studies have been limited to English or number recognition, the recognition range has been expanded through additional data training on Korean text. As a result, this study derived a deep learning-based character recognition algorithm for Korean text recognition. The algorithm obtained a score of 0.841 on the 1-NED evaluation method, which is a similar result to that of English recognition. Further, based on the analysis of the results, major issues with Korean text recognition and possible future study tasks are introduced.