• Journal of the Korea institute for structural maintenance and inspection
• /
• v.27 no.5
• /
• pp.113-119
• /
• 2023

Most infrastructure structures were completed during periods of economic growth. The number of infrastructure structures reaching their lifespan is increasing, and the proportion of old structures is gradually increasing. The functions and performance of these structures at the time of design may deteriorate and may even lead to safety accidents. To prevent this repercussion, accurate inspection and appropriate repair are requisite. To this end, demand is increasing for computer vision and deep learning technology to accurately detect even minute cracks. However, deep learning algorithms require a large number of training data. In particular, label images indicating the location of cracks in the image are required. To secure a large number of those label images, a lot of labor and time are consumed. To reduce these costs as well as increase detection accuracy, this study proposed a learning structure based on mean teacher method. This learning structure was trained on a dataset of 900 labeled image dataset and 3000 unlabeled image dataset. The crack detection network model was evaluated on over 300 labeled image dataset, and the detection accuracy recorded a mean intersection over union of 89.23% and an F1 score of 89.12%. Through this experiment, it was confirmed that detection performance was improved compared to supervised learning. It is expected that this proposed method will be used in the future to reduce the cost required to secure label images.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.28 no.4
• /
• pp.55-61
• /
• 2024

Recently, the number of aging concrete structures is steadily increasing. This is because many of these structures are reaching their expected lifespan. Such structures require accurate inspections and persistent maintenance. Otherwise, their original functions and performance may degrade, potentially leading to safety accidents. Therefore, research on objective inspection technologies using deep learning and computer vision is actively being conducted. High-resolution images can accurately observe not only micro cracks but also spalling and exposed rebar, and deep learning enables automated detection. High detection performance in deep learning is only guaranteed with diverse and numerous training datasets. However, surface damage to concrete is not commonly captured in images, resulting in a lack of training data. To overcome this limitation, this study proposed a method for generating concrete surface damage images, including cracks, spalling, and exposed rebar, using stable diffusion. This method synthesizes new damage images by paired text and image data. For this purpose, a training dataset of 678 images was secured, and fine-tuning was performed through low-rank adaptation. The quality of the generated images was compared according to three base models of stable diffusion. As a result, a method to synthesize the most diverse and high-quality concrete damage images was developed. This research is expected to address the issue of data scarcity and contribute to improving the accuracy of deep learning-based damage detection algorithms in the future.

• Journal of Korean Home Economics Education Association
• /
• v.18 no.1 s.39
• /
• pp.17-29
• /
• 2006

The purpose of this research was to examine the problems, confidence and satisfaction of teachers on the subject ${\ulcorner}technology and home economics{\lrcorner}$ in the 7th national curriculm. For this research, questionnaires were sent by post to teachers who teach technology and home economics in middle schools and high schools. The collected questionnaires were technically analyzed by SPSS/WIN 10.0 program, which measured frequency, percentage, average, standard deviation. According to the types of data, they were also analyzed by t-test and cross tabulation analyses. The results of this research were summarized as follows. 1) There were two teaching types of technology and home economics: the partial charge and the whole charge teaching according to teachers' majors, and both types occurred in similar percentage. The partial charge teaching means that teachers majoring in technology teach only the technology part and teachers majoring in home economics teach only the home economics part when they teach the same subject, technology and home economics. These days the partial charge teaching more often occurs in national or public schools than in private schools, and in coeducational schools than in girls' or boys' schools. 2) The major problems of teaching technology and home economics were caused in order by teachers' lack of skills and knowledge which we not their own major, the lack of students' interests and teaching materials, and burden of tests. 3) Teachers' confidence in teaching the contents of the subject, technology and home economics, made a significant difference according to their majors. Teachers whose major was technology felt more confident when they taught the chapters of the textbooks related to their major, technology, while teachers whose major was home economics felt more confident when they taught the chapters of the textbooks related to their major, home economics. According to implementation types, the partial charge teaching gave higher confidence to the teachers than the whole charge one in teaching almost all the chapters of the textbook. 4) According to implementation types, teachers' satisfaction was showed to be higher in the partial charge teaching than in the whole charge one.

• Journal of Korean Home Economics Education Association
• /
• v.19 no.1 s.43
• /
• pp.81-97
• /
• 2007

This study investigated the Present status of laboratories for experiment and practice, and analyzed teachers' recognitions on experiment and practice for Home Economics area of a Technology and Home Economics curriculum according to majors of teachers. Questionnaires were mailed to middle school teachers who taught home economics part and they answered on the web. 220 replies were used for the final analysis. The findings were as follows: First, the facilities and teaching equipments of laboratories for home economics area were inferior, especially, for clothing and textiles part and housing part. Second, teachers recognized necessity to conduct experiment and practice highly. Food life part scored the highest, while housing part scored the lowest. Teachers who majored in home economics recognized more necessities of experiment and practice than teachers who didn't majored in home economics. Third, they recognized level of experiment and practice to be suitable to students, but 'maintenance and repair of housing' section was relatively less suitable than other sections. Fourth, 'making clothes and recycling' section was recognized to have the least suitability in quantities and hours of experiment and practice lesson, because of too much contents and lack of lesson hours. Fifth, teachers recognized that students were more interested in 'the basis of food preparation and practice' section, but they are less interested in 'maintenance and repair of housing' section. Sixth, teachers recognized that contents of experiment and practice were very useful to the real life. 'The basis of food preparation and practice' section was the most useful, while 'maintenance and repair of housing' section was the least useful. Seventh, experiment and practice lessons for food life part were put in practice very well, followed by the order of clothing and textiles part and housing part. Teachers who majored in home economics usually took more experiment and practice lessons than teachers who didn't major in home economics.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.4
• /
• pp.144-159
• /
• 2014

Techniques for making a single panoramic image using multiple pictures are widely studied in many areas such as computer vision, computer graphics, etc. The panorama image can be applied to various fields like virtual reality, robot vision areas which require wide-angled shots as an useful way to overcome the limitations such as picture-angle, resolutions, and internal informations of an image taken from a single camera. It is so much meaningful in a point that a panoramic image usually provides better immersion feeling than a plain image. Although there are many ways to build a panoramic image, most of them are using the way of extracting feature points and matching points of each images for making a single panoramic image. In addition, those methods use the RANSAC(RANdom SAmple Consensus) algorithm with matching points and the Homography matrix to transform the image. The SURF(Speeded Up Robust Features) algorithm which is used in this paper to extract featuring points uses an image's black and white informations and local spatial informations. The SURF is widely being used since it is very much robust at detecting image's size, view-point changes, and additionally, faster than the SIFT(Scale Invariant Features Transform) algorithm. The SURF has a shortcoming of making an error which results in decreasing the RANSAC algorithm's performance speed when extracting image's feature points. As a result, this may increase the CPU usage occupation rate. The error of detecting matching points may role as a critical reason for disqualifying panoramic image's accuracy and lucidity. In this paper, in order to minimize errors of extracting matching points, we used $3{\times}3$ region's RGB pixel values around the matching points' coordinates to perform intermediate filtering process for removing wrong matching points. We have also presented analysis and evaluation results relating to enhanced working speed for producing a panorama image, CPU usage rate, extracted matching points' decreasing rate and accuracy.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.10
• /
• pp.163-169
• /
• 2017

The purpose of this study is to present the basic data needed to improve the arts, music, and physical education provided by local child care centers based on an investigation of the current status of and satisfaction with the education. The subjects of this study were 17 local child care centers in Gyeonggi-do, South Korea, and the situation of the arts, music, and physical education programs operated from 2014 to 2016 were examined. In addition, 419 children who received the education were surveyed to evaluate their level of satisfaction. The results of this study are as follows. As regards the status of the arts education from 2014 to 2016, it was observed that three of the 17 local child care centers did not have any arts, music or physical education at all, while six of them continuously implemented all three of these programs during this period of time. Two and six of the 17 institutes had arts, music, and physical education programs for two years and one year, respectively. All of the teachers who ran the arts and music education programs of the 17 institutes were arts and music majors who were certified teachers of the liberal arts. However, the physical education programs were run as volunteer activities by college students majoring in physical education. The survey on the level of satisfaction of the children who participated in the arts, music, and physical education programs showed that they were helpful for the overall life experience of the children and that they were more helpful for the boys than for the girls. The level of satisfaction with the education was high for most of the students who participated in the programs, however the boys were more satisfied than the girls. When asked whether they would participate in the arts, music, and physical education programs again, most of the respondents answered that they would do so. The boys were more likely to participate again than the girls. Based on this study, in order to enhance the creativity and personality education of the children using the local child care centers, higher quality education is needed. Arts and music education can be used to help children to learn to communicate smoothly with their friends. In addition, it seems to be necessary to enhance the education by setting goals that are suitable for its purpose, in order to provide creative arts and music education that contributes to the physical health and emotional stability of the children.

• Tuberculosis and Respiratory Diseases
• /
• v.61 no.6
• /
• pp.547-553
• /
• 2006

Background: Liquid-based cytology is currently known as an effective method, and cervical cytology has been shown to be especially effective from of malignancy detection. In our study, the cytological detection rates of the Thinprep (Liquid-based cytology) and conventional cytology (bronchial washing & brushing) for endobronchial lesions were compared. Methods: Between July 2005 and September 2005, the data from 30 patients with respiration symptom, who had shown abnormal lesion on bronchoscopy, were collected. Results: The bronchoscopic biopsy group was consisted of 30 cytodiagnosis specimens, 24 of which were confirmed to be malignant. The others were tuberculosis (4), bronchiectasis and bronchopulmonary fistula (1 each). Of the 24 malignant case, cancer or atypical cells were detected in 19, 17 and 12 of the Thinprep, brushing cytology and washing cytology cases, respectively. None one of the methods detected cancer cells in the non-malignant specimens. Washing cytology has shown sensitivity, specificity, and positive and negative predictive values of 50, 100, 100 and 33.3% respectively. Brushing cytology has shown sensitivity, specificity, and positive and negative predictive values of 70.8, 100, 100 and 46.2%, respectively. Thinprep has shown sensitivity, specificity, and positive and negative predictive values of 79.2, 100, 100 and 54%, respectively. Conclusions: Thinprep (liquid-based cytology) showed better sensitivity and negative predictive values for the evaluation of lung cancer than conventional cytology. However a large-scale study will be needed in the future.

• The Journal of the Korea Contents Association
• /
• v.8 no.12
• /
• pp.291-300
• /
• 2008

We survey recognition of nutrition labeling on the processed foods for the college students with distinction of sex and their major. The frequency of purchase that processed food is over the 3-5 times in a week. The man students preferred to have Ramyon, milk and milk products and the woman students' ingested snacks, soft drinks and sugar snacks. For another, the food major students like to have juice, soft drink, milk and milk products. Then again, the non-food major students ingested Ramyon, snacks and sugar snacks. The woman students and food major students show higher recognition of nutrition labeling and confirmation of it. The man student replied reason why to confirm nutrition labeling is to keep their health. But the woman students show interest to confirm nutrition content. The food major students confirm the nutrition labeling to determine the nutrition labels. The non-food major students did not confirm the nutrition labeling because they think it is an involved style. For knowledge of nutrition contents, the woman student and the student majoring food are well informed. But, all of the student show poor knowledge for staple foods, nutrient function and vitamin. As the research results, we suggest that the educated the student nutrition knowledge for nutrition labeling on the products. It helps to improve their dietary life and eating habits. And they can avoid buying of the processed foods by habit without confirmation of the nutrition.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.49 no.3
• /
• pp.10-17
• /
• 2012

In general, image-based 3D scenes can now be found in many popular vision systems, computer games and virtual reality tours. In this paper, we propose a method for creating 3D virtual scenes based on 2D image that is completely automatic and requires only a single scene as input data. The proposed method is similar to the creation of a pop-up illustration in a children's book. In particular, to estimate geometric structure information for 3D scene from a single outdoor image, we apply the tensor voting to an image segmentation. The tensor voting is used based on the fact that homogeneous region in an image is usually close together on a smooth region and therefore the tokens corresponding to centers of these regions have high saliency values. And then, our algorithm labels regions of the input image into coarse categories: "ground", "sky", and "vertical". These labels are then used to "cut and fold" the image into a pop-up model using a set of simple assumptions. The experimental results show that our method successfully segments coarse regions in many complex natural scene images and can create a 3D pop-up model to infer the structure information based on the segmented region information.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.8
• /
• pp.87-96
• /
• 2014

In this paper, we propose a directionally adaptive aliasing and noise removal using dictionary learning based on space-frequency analysis. The proposed aliasing and noise removal algorithm consists of two modules; i) aliasing and noise detection using dictionary learning and analysis of frequency characteristics from the combined wavelet-Fourier transform and ii) aliasing removal with suppressing noise based on the directional shrinkage in the detected regions. The proposed method can preserve the high-frequency details because aliasing and noise region is detected. Experimental results show that the proposed algorithm can efficiently reduce aliasing and noise while minimizing losses of high-frequency details and generation of artifacts comparing with the conventional methods. The proposed algorithm is suitable for various applications such as image resampling, super-resolution image, and robot vision.