• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.79-104
• /
• 2020

Recently, as deep learning has attracted attention, the use of deep learning is being considered as a method for solving problems in various fields. In particular, deep learning is known to have excellent performance when applied to applying unstructured data such as text, sound and images, and many studies have proven its effectiveness. Owing to the remarkable development of text and image deep learning technology, interests in image captioning technology and its application is rapidly increasing. Image captioning is a technique that automatically generates relevant captions for a given image by handling both image comprehension and text generation simultaneously. In spite of the high entry barrier of image captioning that analysts should be able to process both image and text data, image captioning has established itself as one of the key fields in the A.I. research owing to its various applicability. In addition, many researches have been conducted to improve the performance of image captioning in various aspects. Recent researches attempt to create advanced captions that can not only describe an image accurately, but also convey the information contained in the image more sophisticatedly. Despite many recent efforts to improve the performance of image captioning, it is difficult to find any researches to interpret images from the perspective of domain experts in each field not from the perspective of the general public. Even for the same image, the part of interests may differ according to the professional field of the person who has encountered the image. Moreover, the way of interpreting and expressing the image also differs according to the level of expertise. The public tends to recognize the image from a holistic and general perspective, that is, from the perspective of identifying the image's constituent objects and their relationships. On the contrary, the domain experts tend to recognize the image by focusing on some specific elements necessary to interpret the given image based on their expertise. It implies that meaningful parts of an image are mutually different depending on viewers' perspective even for the same image. So, image captioning needs to implement this phenomenon. Therefore, in this study, we propose a method to generate captions specialized in each domain for the image by utilizing the expertise of experts in the corresponding domain. Specifically, after performing pre-training on a large amount of general data, the expertise in the field is transplanted through transfer-learning with a small amount of expertise data. However, simple adaption of transfer learning using expertise data may invoke another type of problems. Simultaneous learning with captions of various characteristics may invoke so-called 'inter-observation interference' problem, which make it difficult to perform pure learning of each characteristic point of view. For learning with vast amount of data, most of this interference is self-purified and has little impact on learning results. On the contrary, in the case of fine-tuning where learning is performed on a small amount of data, the impact of such interference on learning can be relatively large. To solve this problem, therefore, we propose a novel 'Character-Independent Transfer-learning' that performs transfer learning independently for each character. In order to confirm the feasibility of the proposed methodology, we performed experiments utilizing the results of pre-training on MSCOCO dataset which is comprised of 120,000 images and about 600,000 general captions. Additionally, according to the advice of an art therapist, about 300 pairs of 'image / expertise captions' were created, and the data was used for the experiments of expertise transplantation. As a result of the experiment, it was confirmed that the caption generated according to the proposed methodology generates captions from the perspective of implanted expertise whereas the caption generated through learning on general data contains a number of contents irrelevant to expertise interpretation. In this paper, we propose a novel approach of specialized image interpretation. To achieve this goal, we present a method to use transfer learning and generate captions specialized in the specific domain. In the future, by applying the proposed methodology to expertise transplant in various fields, we expected that many researches will be actively conducted to solve the problem of lack of expertise data and to improve performance of image captioning.

• Imaging Science in Dentistry
• /
• v.40 no.1
• /
• pp.25-32
• /
• 2010

Purpose : This study was designed to evaluate the efficacy and utility of panoramic radiograph for the improvement of the periodic oral examinations. Materials and Methods : Clinical examinations and panoramic examinations were done for the 242 subjects of oral examinations. The results of panoramic radiograph interpretation were compared with the clinical findings. Two questionnaires were created. One was carried out before the panoramic examination and the other done afterwards, to find out the subjects' cognition and satisfaction for the clinical and panoramic examinations. Results : 1. Panoramic findings showed a higher detection rate of 31.9% for periodontal diseases, and 23.1% for dental caries than clinical findings. 2. The additional abnormalities detected through panoramic examinations were impacted tooth in 81 subjects (33.6%), maxillary sinus abnormalities in 28 subjects (11.6%), condylar abnormalities in 5 subjects (2.1%), congenital and acquired dental anormalies in 59 subjects (24.5%), and other miscellaneous abnormalities in 34 subjects (14.1%). 3. 164 subjects (67.8%) were satisfied with the current periodic oral examination, and 75 subjects (31.1%) hoped for better accuracy. 4. In the first and second questionnaire, 154 subjects (67.0%) and 163 subjects (70.6%) responded respectively that panoramic examination was necessary, and 193 subjects (83.2%) responded that it actually helped. Conclusion : The panoramic examination was revealed to improve the effectiveness of the periodic oral examination and to increase the satisfaction of the subjects of examination.

• International Journal of Advanced Culture Technology
• /
• v.5 no.3
• /
• pp.11-23
• /
• 2017

Anatomy is one of the basic subjects of the nursing course, which is included in the curriculum of the nursing. Anatomy is a basic course for understanding major in nursing and it is the first gateway to acquire expertise. It is mainly opened in nursing and first to second grades. Therefore, students who have advanced to the nursing department have great interest and expectation on the anatomy subjects. In general, nursing students are studying anatomy with theories and models, and some universities practice on tour after dissection of medical consortium for short time. This is called the Cadaver practice. This study was carried out to investigate the thoughts and experiences of bioethics through nursing students' Cadaver practice. The interview data were processed through the analysis and interpretation process using the phenomenological research method, Giorgi method. As a result, 48 semantic units were derived, and then they were divided into 10 subcomponents and divided into 6 categories. As a result of the analysis, the experience of nursing students' bioethics was tied to the topic of six questions. For example, 'what about the respect and responsibility of nursing students in Cadaver?', 'What about your experience with the Cadaver experiment and bioethics?', 'What was the academic achievement of the actual human body structure viewing experience with cadaver?', 'What was the connection with theory?', 'What was your intention to recommend to others?', 'What was your perception and interest in the Cadaver experiment?'. Analyzes were integrated into 10 structures; "Thank you for your donation", "want others to refuse donation", "Burden of practice", "Good opportunities for learning", "Understand better", "Should study harder", "Memory is better", "Compared to theory", "Good experience", "Want to rejoin". The general structural description of the participants' meanings is summarized as follows. Nursing college students who participated in the Cadaver had a gratitude for the donor, but they said they would like to talk to others about donating organs. Before they went to practice, they felt a lot of pressure on the dissection of the cadaver, but they went to the practice and thought that it was a good opportunity to learn by doing well. Specifically, they understood that they had better understanding than the theoretical lesson, and that they were more eager to study their major through practice. In other words, most of them were more memorable and they would like to participate again if the opportunity comes next time. The results of this study show that the practice of Cadaver in nursing college students is very positive in terms of educational effectiveness. However, in terms of bioethics, it can be seen that the education process is somewhat unsatisfactory. Therefore, the systematic bioethics education should be prepared before the practice in the nursing college students' Cadaver practice.

• Journal of the Korean Institute of Gas
• /
• v.23 no.4
• /
• pp.46-64
• /
• 2019

Technical practice code, KGS GC101 2018, for explosion hazard area selection and distance calculation of gas facility was enacted and implemented from July 12, 2018. This code includes whole contents of IEC60079-10-1 2015 (Explosive atmospheres Part 10-1: Classification of areas - Explosive gas atmospheres), and clarifies the interpretation of ambiguous standards or adds guidelines for standards. KGS GC101 is a method for classifying explosion hazard place types: (1) Determination of leak grade (2) Determination of leakage hole size (3) Determination of leakage flow (4) Determination of dilution class (5) Determination of ventilation effectiveness, finally (6) Determination of danger place (7) Explosion The range of dangerous places can be estimated. In order to easily calculate this process, the program (KGS-HAC v1.14, C-2018-020632) composed by Visual Basic for Application (Excel) language was produced by Korea Gas Safety Corporation. We will discuss how to use codes and programs to select and set up explosion hazard zones for field users.

• The Korean Society of Law and Medicine
• /
• v.13 no.2
• /
• pp.11-38
• /
• 2012

The Supreme Court stand in the position in specific lawsuit that it doesn't allow the discretionary not covered service, but recently in revocation suit of fine disposal that is imposed on medical fee of leukemia patient, it altered the existing adjudgement and admitted the discretionary not covered service exceptionally. It put forward the allowable condition roughly in that case. According as this alteration, it has become more important to embody the allowance conditions of exceptions. The Supreme Court presented three things, which are procedural condition, medical condition and subscriber's agreement. Concerning procedural condition, several present conciliation procedures are as follows: medical care benefit arret request, relative value conciliation etc, prior request on anti-cancer drug among chemicals which exceed acceptance criteria, request of non benefit object on common drugs. To be granted the existence of those system, there should be no obstacle to use that. Even if it were so, we should take circumstances into consideration; individual situation is unescapable concerning substance and urgency of the discretionary not covered service, process of the procedure, time required etc. Regarding medical condition, safety and effectiveness will be verified through evaluation procedures of new medical skill. About the necessity, the Supreme Court made clear through a sentence that it allow the discretionary not covered service, in case that needs to treat a patient out of the standard of medical benefit. Strict interpretation is right and it answer the purpose of the sentence that the supreme court permit the discretionary not covered service, exceptionally. We need to differentiate medical necessity and medical validity. Subscriber's agreement should holds true if it entails full explanation, and if it is preliminary, explicit and individual. On this account, it should be difficult to admit that someone agree effectively when he call for the affirmation that he is recipient of medical care. Reasonable expense needs to be a part of review whether the agreement is valid. Meanwhile If we adjust system of medical expense and eventually reorganize a fee for consultation payment system (Fee-for-service controlled by item to DRG (Diagnosis Related Groups)), controversial area of the discretionary not covered service will be decreased and that will guarantee the discretion of the doctor.

• The Journal of the Korea Contents Association
• /
• v.21 no.3
• /
• pp.626-633
• /
• 2021

In the event of a housing sale guarantee accident, the Housing and Urban Guarantee Corporation shall be responsible for the guarantee by refunding the moving-in money unless there are special circumstances to the contractor. The Housing and Urban Guarantee Corporation may refuse to perform the refund based on the Terms and Conditions Regulations Act, but disputes continue to arise between interested parties in this regard. Therefore, the purpose of this study is to study the problems and improvement measures for legal issues related to the implementation of the refund. First, the issue of guarantee effectiveness and scope of guarantee under the Terms and Conditions Regulation Act. Second, the problem of guarantee contract for conditional third parties of Housing and Urban Guarantee Corporation. Third, the problem of the attitude of the existing precedents of the Supreme Court was examined. As a result of reviewing these legal issues, it was confirmed that the interpretation of terms and conditions according to the implementation of the refund is being interpreted in accordance with the principle of good faith, but according to individual cases, precedents can be divided into positive and negative judgments. In addition, despite the fact that the housing pre-sale guarantee is a guarantee contract for a conditional third party, it was confirmed that the buyer suffered damages in good faith through active disputes with the interested parties. Accordingly, the Housing and Urban Guarantee Corporation proposed an improvement plan for roles and cooperation items to meet the purpose of establishment for the stability of the customer's housing.

• Journal of the Korean Geotechnical Society
• /
• v.39 no.10
• /
• pp.41-48
• /
• 2023

The presence of abnormalities in the subgrade of roads poses safety risks to users and results in significant maintenance costs. In this study, we aimed to experimentally evaluate the temperature distributions in abnormal areas of subgrade materials using infrared cameras and analyze the data with machine learning techniques. The experimental site was configured as a cubic shape measuring 50 cm in width, length, and depth, with abnormal areas designated for water and air. Concrete blocks covered the upper part of the site to simulate the pavement layer. Temperature distribution was monitored over 23 h, from 4 PM to 3 PM the following day, resulting in image data and numerical temperature values extracted from the middle of the abnormal area. The temperature difference between the maximum and minimum values measured 34.8℃ for water, 34.2℃ for air, and 28.6℃ for the original subgrade. To classify conditions in the measured images, we employed the image analysis method of a convolutional neural network (CNN), utilizing ResNet-101 and SqueezeNet networks. The classification accuracies of ResNet-101 for water, air, and the original subgrade were 70%, 50%, and 80%, respectively. SqueezeNet achieved classification accuracies of 60% for water, 30% for air, and 70% for the original subgrade. This study highlights the effectiveness of CNN algorithms in analyzing subgrade properties and predicting subsurface conditions.

• Journal of Korea Entertainment Industry Association
• /
• v.13 no.5
• /
• pp.145-153
• /
• 2019

The present study is a discussion in which the flow of 'social and cultural values' inherent in the creative advertising story is considered against Bart's symbolism and the creative symbol system, and attempted to reproduce the work through the cognitive thinking of the inmates. The interaction of correct social and cultural communication is not just a strategy for persuasion and effectiveness. Starting with these issues, I thought that experiencing the 'symbolic production' and 'cognition interpretation' of the most creative, aesthetic and implicit advertising stories was the realization of concrete cultural values. The reason why I pay attention to advertising as a target tool of the original school is that it gives anyone access to the social and cultural values based on the productivity of meaning, the sharing of meaning and social small-call work by paying attention to the most implicit symbols in a short period of time. I also think that with the trend of the times, it is well worth it as a tool of positive communication for social and cultural member harmony and solving future problems. The reality of social and cultural advertising stories conducted in conjunction with the analysis of meaning at the cognitive thought level is very appropriate to apply in creative classes for college students. The Dong-A Ilbo is a discussion that suggested that the work of realizing the cognitive meaning of advertising stories, a "symbol complex" based on creativity in a complex, multi-media era, will become an age-old communication tool to join university students' strategies for solving future problems

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.17 no.1
• /
• pp.1-6
• /
• 2011

The underlying principle of environmental conservation and restoration is to receive continuous services from nature. In this context, the basic objective of oil spill response and recovery activities is to allow affected residents to retain sustainability of their daily livelihood and productive environment. Hence, monitoring the status of socio-economic impact and the extent of restoration is an important parameter in the restoration effort to repair the damaged functions of environmental services with oil spill accident. However, assessment of socio-economic impact poses investigators with much difficulties in selecting the most appropriate approaches because the process often involves various stakeholders, directly and indirectly. Moreover, the analyses and interpretation of the results also pose a great challenge. The present study monitored fisheries and tourism numbers which were considered as major local socio-economic indicators of living and production environment affected by M/T Hebei Spirit oil-spill accident. This monitoring was conducted by examining the published papers and statistical reports. This was supplemented by surveying how the local residents actually felt about the damage and recovery for the first time in Korea. The results showed that the rate of the recovery was about 40~50 %, and the rate of the recovery seemed to be slowed or decreased. However, what the local residents actually felt was 2~10 points less than the literature surveys and statistical reports. These results suggested limits to using only the literature and statistical surveys for the traditional socio-economic impact assessment. The study also showed the need to include in the impact assessment process what and how the local residents actual feel about the oil spill damage and recovery process.

• Geophysics and Geophysical Exploration
• /
• v.10 no.1
• /
• pp.50-59
• /
• 2007

The advent of towed geo-electrical array surveying on water and land has resulted in datasets of magnitude approaching that of airborne electromagnetic surveying and most suited to 1D inversion. Robustness and complete automation is essential if processing and reliable interpretation of such data is to be viable. Sharp boundaries such as river beds and the top of saline aquifers must be resolved so use of smoothness constraints must be minimised. Suitable inversion algorithms must intelligently handle low signal-to-noise ratio data if conductive basement, that attenuates signal, is not to be misrepresented. A noise-level aware inversion algorithm that operates with one elastic thickness layer per electrode configuration has been coded. The noise-level aware inversion identifies if conductive basement has attenuated signal levels so that they are below noise level, and models conductive basement where appropriate. Layers in the initial models are distributed to span the effective depths of each of the geo-electric array quadrupoles. The algorithm works optimally on data collected using geo-electric arrays with an approximately exponential distribution of quadrupole effective depths. Inversion of data from arrays with linear electrodes, used to reduce contact resistance, and capacitive-line antennae is plausible. This paper demonstrates the effectiveness of the algorithm using theoretical examples and an example from a salt interception scheme on the Murray River, Australia.