• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.413-415
• /
• 2022

Since human joints can be used as useful information for analyzing human behavior as a component of the human body, many studies have been conducted on human action recognition using joint information. However, it is a very complex problem to recognize human action that changes every moment using only each independent joint information. Therefore, an additional information extraction method to be used for learning and an algorithm that considers the current state based on the past state are needed. In this paper, we propose a human action recognition technique considering the positional relationship of connected joints and the change of the position of each joint over time. Using the pre-trained joint extraction model, position information of each joint is obtained, and bone information is extracted using the difference vector between the connected joints. In addition, a simplified neural network is constructed according to the two types of inputs, and spatio-temporal features are extracted by adding LSTM. As a result of the experiment using a dataset consisting of 9 behaviors, it was confirmed that when the action recognition accuracy was measured considering the temporal and spatial relationship features of each joint, it showed superior performance compared to the result using only single joint information.

• Smart Media Journal
• /
• v.11 no.1
• /
• pp.46-57
• /
• 2022

In this paper, multi-region based Radial Graph Convolutional Network (MRGCN) algorithm which can perform end-to-end action recognition using the optical flow and gradient of input image is described. Because this method does not use information of skeleton that is difficult to acquire and complicated to estimate, it can be used in general CCTV environment in which only video camera is used. The novelty of MRGCN is that it expresses the optical flow and gradient of the input image as directional histograms and then converts it into six feature vectors to reduce the amount of computational load and uses a newly developed radial type network model to hierarchically propagate the deformation and shape change of the human body in spatio-temporal space. Another important feature is that the data input areas are arranged being overlapped each other, so that information is not spatially disconnected among input nodes. As a result of performing MRGCN's action recognition performance evaluation experiment for 30 actions, it was possible to obtain Top-1 accuracy of 84.78%, which is superior to the existing GCN-based action recognition method using skeleton data as an input.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.3
• /
• pp.909-914
• /
• 2024

This study suggests a quantitative evaluation of transfer learning, which is widely used in various AI fields, including image recognition for robot vision. Quantitative and qualitative analyses of results applying transfer learning are presented, but transfer learning itself is not discussed. Therefore, this study proposes a quantitative evaluation of transfer learning itself based on MNIST, a handwritten digit database. For the reference network, the change in recognition accuracy according to the depth of the transfer learning frozen layer and the ratio of transfer learning data and pre-training data is tracked. It is observed that when freezing up to the first layer and the ratio of transfer learning data is more than 3%, the recognition accuracy of more than 90% can be stably maintained. The transfer learning quantitative evaluation method of this study can be used to implement transfer learning optimized according to the network structure and type of data in the future, and will expand the scope of the use of robot vision and image analysis AI in various environments.

• Journal of Agricultural Extension & Community Development
• /
• v.6 no.2
• /
• pp.85-103
• /
• 1999

The objectives of the study were to review the agricultural environment education program for nasal change agents, to find out their recognition on agricultural environment and to get sorn implications for better educational program of rural environment. The data were gathered from 177 rural change agents in 11 agricultural technology centers in Kyongi-Do province by mailing survey with questionnaire. The major findings of the study were ; (1) Agricultural environment education for rural extention workers initiated in 1996. From 1999 diversified programs were served for them. (2) Rural change agents perceived that water contamination and degradation of life environment were the most serious problem. (3) Educational program on organic farming should be prepared for farmers and rural changes agents. (4) Concrete educational programs for adequate use of agricultural chemicals should be prepared for farmers. (5) Educational program on agricultural environment policies should be reorganized in terms of the level of target group. (6) RDA should prepare synthetical education program for the conservation of rural eco-system.

• 한국유가공학회:학술대회논문집
• /
• 2006.06a
• /
• pp.47-56
• /
• 2006

The common recognition which the organic product is compared as safe to baby does not mean there is especially plentiful nutrition in organic product. The part mineral content of agricultural products follows in soil quality, therefore plentiful chemical fertilizer-use to change soil quality results in the possibility of the mineral content change. The vitamin, carbohydrate, protein and fatty acid composition does not change largely from original nutrition facts. Rather than, the most advantage of organic product is relieving baby from agricultural chemical harms. It will be a strong desire for consumers to eat safe, believable and healthy food like organic foods. Furthermore, it will be sincerely good to environment and ecosystem.

• Educational Technology International
• /
• v.10 no.2
• /
• pp.107-125
• /
• 2009

This study seeks to prove three points through the research. The first point is to examine the changes of international attitudes with actual experiences using synchronous international distance learning. The second point is to examine the effectiveness of a synchronous international distance system. And the final point is to compare international attitudes among middle school and undergraduate school students in Korea and Japan. The study used the DVTS for audio and video communication tools and automatic translating chat as a text communication tool. This combination of communication tools was very effective for students from both countries to communicate for international collaborative learning activities. The study found several interesting patterns of attitude change from the results of the study. For whole category analysis, there are positive changes in four categories of international attitudes: consciousness to foreign countries, consideration for other's viewpoints, motivation for international education, and recognition for the counterpart country. However, there was no change in the nationality category.

• Proceedings of the Korea Contents Association Conference
• /
• 2006.11a
• /
• pp.801-805
• /
• 2006

In this paper, we describe a system of controlling the pictorial puzzle program using information of hand shape. We extract hand region using skin color information and then principal component analysis uses centroidal profile information which comes blob of 2D appearance for hand shape recognition. This method suit hand shape recognition in real time because it extracts hand region accurately, has little computation quantity, and is less sensitive to lighting change using skin color information in complicated background. Finally, we controlled a pictorial puzzle with using recognized hand shape information. This method has good result when we make an experiment on application of pictorial puzzle. Besides, it can use so many HCI field.

• Proceedings of the KSPS conference
• /
• 2002.11a
• /
• pp.81-85
• /
• 2002

We present a statistical analysis of Korean phonological variations using automatic generation of phonetic transcription. We have constructed the automatic generation system of Korean pronunciation variants by applying rules modeling obligatory and optional phonemic changes and allophonic changes. These rules are derived from knowledge-based morphophonological analysis and government standard pronunciation rules. This system is optimized for continuous speech recognition by generating phonetic transcriptions for training and constructing a pronunciation dictionary for recognition. In this paper, we describe Korean phonological variations by analyzing the statistics of phonemic change rule applications for the 60,000 sentences in the Samsung PBS(Phonetic Balanced Sentence) Speech DB. Our results show that the most frequently happening obligatory phonemic variations are in the order of liaison, tensification, aspirationalization, and nasalization of obstruent, and that the most frequently happening optional phonemic variations are in the order of initial consonant h-deletion, insertion of final consonant with the same place of articulation as the next consonants, and deletion of final consonant with the same place of articulation as the next consonants. These statistics can be used for improving the performance of speech recognition systems.

• Journal of Broadcast Engineering
• /
• v.21 no.1
• /
• pp.51-59
• /
• 2016

This paper presents a method of DCT-based illuminant compensation to enhance the accuracy of face recognition under an illuminant change. The basis of the proposed method is that the illuminant is generally located in low-frequency components in the DCT domain. Therefore, the effect of the illuminant can be compensated by controlling the low-frequency components. Moreover, a directional high-order local pattern descriptor is used to detect robust features in the case of face motion. Experiments confirm the performance of the proposed algorithm got up to 95% when tested using a real database.

• Journal of radiological science and technology
• /
• v.41 no.3
• /
• pp.255-260
• /
• 2018

Most patients and parents and guardians display frequent anxiety due to radiation exposure during outpatient, ward, and pediatric general radiographic examinations. This is a behavior that perceives only the harmfulness of radiation. For the recognition of medical radiation, we conduct surveys on outpatients, inpatients, and pediatric parents and guardians to identify their awareness, and then use the radiation dose promotional materials After providing accurate information on the use of radiation, the outpatient, inpatient, and pediatric parents and guardians were asked to explain the change in awareness. The questionnaire items were classified into five categories: repetitive radiation awareness for diagnosis, awareness of exposure dose, availability of exposure information, awareness of radiation risk, and awareness of health problems caused by radiation. There was a statistically significant difference in the items of recognition result of medical radiation, although there was a slight difference in the individual items in the pre and post-recognition results of providing information about the radiologists of the protector and the outpatient(p<0.05). Therefore, through the installation of these promotional materials, we will improve our awareness of medical radiation safety during general radiography surveillance in the Department of Radiology to provide better quality medical information and medical services, thereby contributing to strengthening the competitiveness of the hospital.