• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.7
• /
• pp.757-763
• /
• 2019

Youtube has become so popular that it is called the age of YouTube. As the number of users and contents increase, the choice of information increases. However, it is difficult to select information that meets the needs of users. YouTube provides recommendations based on their watch list. Therefore, in this study, we want to analyze the channel of user's subject in various angles and provide the proposed scheme based on the crawled channels, measurement of the perception of channels and channel videos through quantitative data and hidden qualitative data analysis. Based on the above two data analysis, it is possible to know the recognition of the channel and the recognition of the channel video, thereby providing a ranking of the channels that deal with the topic. Finally, as a case study, we recommend English learning channels to users based on numerical data statistics and emotional analysis results to maximize flipped learning effect regardless of time and space.

• The Korean Journal of Applied Statistics
• /
• v.32 no.1
• /
• pp.41-68
• /
• 2019

Multi-view learning considers data from various viewpoints as well as attempts to integrate various information from data. Multi-view learning has been studied recently and has showed superior performance to a model learned from only a single view. With the introduction of deep learning techniques to a multi-view learning approach, it has showed good results in various fields such as image, text, voice, and video. In this study, we introduce how multi-view learning methods solve various problems faced in human behavior recognition, medical areas, information retrieval and facial expression recognition. In addition, we review data integration principles of multi-view learning methods by classifying traditional multi-view learning methods into data integration, classifiers integration, and representation integration. Finally, we examine how CNN, RNN, RBM, Autoencoder, and GAN, which are commonly used among various deep learning methods, are applied to multi-view learning algorithms. We categorize CNN and RNN-based learning methods as supervised learning, and RBM, Autoencoder, and GAN-based learning methods as unsupervised learning.

• Journal of Animal Science and Technology
• /
• v.65 no.3
• /
• pp.519-534
• /
• 2023

Rumination in cattle is closely related to their health, which makes the automatic monitoring of rumination an important part of smart pasture operations. However, manual monitoring of cattle rumination is laborious and wearable sensors are often harmful to animals. Thus, we propose a computer vision-based method to automatically identify multi-object cattle rumination, and to calculate the rumination time and number of chews for each cow. The heads of the cattle in the video were initially tracked with a multi-object tracking algorithm, which combined the You Only Look Once (YOLO) algorithm with the kernelized correlation filter (KCF). Images of the head of each cow were saved at a fixed size, and numbered. Then, a rumination recognition algorithm was constructed with parameters obtained using the frame difference method, and rumination time and number of chews were calculated. The rumination recognition algorithm was used to analyze the head image of each cow to automatically detect multi-object cattle rumination. To verify the feasibility of this method, the algorithm was tested on multi-object cattle rumination videos, and the results were compared with the results produced by human observation. The experimental results showed that the average error in rumination time was 5.902% and the average error in the number of chews was 8.126%. The rumination identification and calculation of rumination information only need to be performed by computers automatically with no manual intervention. It could provide a new contactless rumination identification method for multi-cattle, which provided technical support for smart pasture.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.3
• /
• pp.140-147
• /
• 2024

In the Republic of Korea, people with hearing impairments are the second-largest demographic within the registered disability community, following those with physical disabilities. Despite this demographic significance, research on sign language translation technology is limited due to several reasons including the limited market size and the lack of adequately annotated datasets. Despite the difficulties, a few researchers continue to improve the performacne of sign language translation technologies by employing the recent advance of deep learning, for example, the transformer architecture, as the transformer-based models have demonstrated noteworthy performance in tasks such as action recognition and video classification. This study focuses on enhancing the recognition performance of sign language translation by combining transformers with 3D-CNN. Through experimental evaluations using the PHOENIX-Wether-2014T dataset [1], we show that the proposed model exhibits comparable performance to existing models in terms of Floating Point Operations Per Second (FLOPs).

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.10
• /
• pp.47-57
• /
• 2024

This paper proposes a system to efficiently manage delays caused by unmanaged and congested queues in crowded environments. Such queues not only cause inconvenience but also pose safety risks. Existing systems, relying on single-camera feeds, are inadequate for complex scenarios requiring multiple cameras. To address this, we developed a multi-vision long queue detection system that integrates multiple vision algorithms to accurately detect various types of queues. The algorithm processes real-time video data from multiple cameras, stitching overlapping segments into a single panoramic image. By combining object detection, tracking, and position variation analysis, the system recognizes long queues in crowded environments. The algorithm was validated with 96% accuracy and a 92% F1-score across diverse settings.

• Korean Journal of Health Education and Promotion
• /
• v.18 no.1
• /
• pp.109-125
• /
• 2001

The purpose of this study was to provide the basic data for establishing school education environment protection measures, on the basis of comparing and analyzing the realities and students' recognition degree of the environment and hygiene around the middle and high schools located in the rural areas. These study data were investigated by the self-administered questionnaires, taking as subject the 805 students in the middle and high schools located rural areas. And the results were as follows: First, as the result of having investigated the distribution degree of harmful environment within the purification zone around schools, it was found out that students responded: within the purification zone around the middle school, there were cartoon rooms (46.2％), electronic game rooms (45.9％), and singing rooms (45.0％). within the purification zone around the high school, there were electronic game rooms (46.3％), singing rooms (42.3％), billiard halls (41.4％), PC rooms (40.1 ％), and Soju-room (35.2％). Secondly, as having analyzed student's recognition degree of the harmful environment around the school, it was found out that middle school students responded that sexual utensils-treating shops (3.74 points) were most harmful, and next corrupted bathhouses (3.52 points), and Soju-room (3.47 points), and high school students also responded relating to harmfulness in a similar sequence. Thirdly, in case of students' recognition degree of the harmful environment around the school according to general characteristics, 1) girl students had a higher ratio of recognition that the environment around the school was harmful than boy students (p〈0.001). 2) groups of students whose living standard was high had a higher ratio of recognition that the environment around the school was harmful than groups of students whose living standard was low (p〈0.05). 3) groups of students whose school was located near the park or the residential street had a higher degree of recognition that the environment around the school was harmful than groups of students whose school was located near the factory or the shopping area (p〈0.01). 4) groups of students whose school was located near the park or the residential street had a higher degree of recognition that the environment around the school was harmful than groups of students whose school was located near the amusement area or the shopping area (p〈0.05). Fourthly, 1) relating to the harmful shops where they experienced most highly the behavior of drinking and smoking, middle school students responded that they did so in the electronic game room (22.5％) and high school students did so in the singing room (31.4％), and high school students had a very high experience ratio of drinking and smoking, compared with middle school students (p〈0.001). 2) relating to the harmful shops where they could get in contact with lewd articles, both of middle school students (5.3％) and high school students (8.3％) responded that they could do so in the video room. 3) relating to the harmful shops where they experienced unsound opposite sex acquaintance, both of middle school students (5.8％) and high school students (16.6％) responded that they did so most highly in hotels, and high school students had a remarkably high experience ratio of unsound opposite sex acquaintance, compared with middle school students (p〈0.05). 4) relating to the harmful shops where they experienced violence, middle school students responded that they did so in the electronic game room (14.0％) and then in the singing room (3.7％), and high school students responded that they did so in the electronic game room (9.3％), the nightclub (4.6％), Soju-room (4.1 ％), and high school students had a remarkably high experience ratio of violence, compared with middle school students (p〈0.05). 5) relating to the harmful places where they experienced drugs both of middle school students (0.8％) and high school students (2.4％) responded that they did so in the hotels. Fifthly, when going to the harmful shops, students had the experience of being guided and regulated roughly 1 time - 2 times, and middle school students (16.4％) and high school students (16.7％) had almost similar experience ratios of being guided and regulated. Conclusively, there was a limit in controlling the environment and purification zone only by legal regulations and institutional controls, the self-control purification effort for the school and the surrounding environment was required greatly, in order to protect students from harmful environment. In addition, the constant study to establish the educational environment purification measures must be carried out.

• Journal of Internet Computing and Services
• /
• v.19 no.6
• /
• pp.41-51
• /
• 2018

Recently, in the field of video surveillance, deep learning based learning method is applied to intelligent video surveillance system, and various events such as crime, fire, and abnormal phenomenon can be robustly detected. However, since occlusion occurs due to the loss of 3d information generated by projecting the 3d real-world in 2d image, it is need to consider the occlusion problem in order to accurately detect the object and to estimate the pose. Therefore, in this paper, we detect moving objects by solving the occlusion problem of object detection process by adding depth information to existing RGB information. Then, using the convolution neural network in the detected region, the positions of the 14 keypoints of the human joint region can be predicted. Finally, in order to solve the self-occlusion problem occurring in the pose estimation process, the method for 3d human pose estimation is described by extending the range of estimation to the 3d space using the predicted result of 2d keypoint and the deep neural network. In the future, the result of 2d and 3d pose estimation of this research can be used as easy data for future human behavior recognition and contribute to the development of industrial technology.

• Journal of Digital Convergence
• /
• v.19 no.3
• /
• pp.203-209
• /
• 2021

In this study, a prototype design model was proposed for developing an online recruitment system through multi-dimensional data crawling and social media analysis, and validates text information and video interview in job application process. This study includes a comparative analysis process through text mining to verify the authenticity of job application paperwork and to effectively hire and allocate workers based on the potential job capability. Based on the prototype system, we conducted performance tests and analyzed the result for key performance indicators such as text mining accuracy and interview STT(speech to text) function recognition rate. If commercialized based on design specifications and prototype development results derived from this study, it may be expected to be utilized as the intelligent online recruitment system technology required in the public and private recruitment markets in the future.

• Journal of Practical Engineering Education
• /
• v.5 no.2
• /
• pp.109-115
• /
• 2013

In this paper, a teaching assistant system using computer vision is presented. Using the proposed system, lecturers can utilize various lecture contents such as lecture notes and related video clips easily and seamlessly. In order to do transition between different lecture contents and control multimedia contents, lecturers just draw pre-defined symbols on the board without pausing the class. In the proposed teaching assistant system, a feature descriptor, so called shape context, is used for recognizing the pre-defined symbols successfully.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.1
• /
• pp.85-92
• /
• 2020

Deep learning shows outstanding performance in image and video analysis, such as object classification, object detection and semantic segmentation. In this paper, it is analyzed that the performances of deep learning models can be affected by characteristics of train dataset. It is proposed as a method for selecting activation function and optimization algorithm of deep learning to classify facial expression. Classification performances are compared and analyzed by applying various algorithms of each component of deep learning model for CK+, MMI, and KDEF datasets. As results of simulation, it is shown that genetic algorithm can be an effective solution for optimizing components of deep learning model.