• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2014.06a
• /
• pp.258-261
• /
• 2014

The era of realistic broadcast with high fidelity has come after the wide-spread distribution of UHD display and the transmission of UHD experimental broadcast in CATV. However, UHD broadcast now has constraint because it requires much amount of bandwidth and data in broadcasting transmission and production system. Not only HEVC(High Efficiency Video Codec) which has more than two times higher compression rate but also cloud-based editing system would be the key to solve the problems above. Also, fast scene change detection of videos is needed to index and search UHD videos smoothly. In this paper, therefore, a method is proposed to index and search the scene change information of large volume UHD videos compressed with high-efficiency codec. Application usages of fast detection of scene change information in various UHD video environments are considered by using this algorithm.

• Journal of the Korea Convergence Society
• /
• v.11 no.8
• /
• pp.39-45
• /
• 2020

Recently the terrors have been attempted in the public places of the nations such as United states, England and Japan by explosive things, toxic materials and so on. It is understood that the method in which dangerous objects are put in public places is one of the difficult types in detection. While there are the cameras recording videos for many spots in public places, it is very hard for the security personnel to monitor every videos. Nowadays the smart softwares which can analyzing videos automatically are utilized to detect abandoned objects. The method by Lin et al. shows comparatively high detection rates for abandoned objects but it is not easy to obtain the shape information because there is a tendency that the number of the pixels decreases abruptly along the time goes due to the characteristics of short-term background images. In this research a novel method is proposed to successfully extract the shape of the abandoned object by analysing the characteristics of area variation. The experiment results show that the proposed method has better performance in extracting shape information in comparison with the precedent approach.

• Journal of Korea Multimedia Society
• /
• v.15 no.5
• /
• pp.577-586
• /
• 2012

The smartphone emerged due to the rapid development of the Internet has brought greater convenience to life in a positive manner. Recently, however, because of unconstrained exposure to harmful video, reckless use of smart phones has become a domestic issue in our society. In this paper, a service which detects harmful videos by using the acoustic information is designed and implemented on the Android OS. In order to implement the service of Android OS-based detection of the harmful movie, the speed of existing sound-based detection method for harmful videos is improved. The GMM(Gaussian Mixture Model) was used for classifier and the number of Gaussian Mixture was 18. The implemented service shows a detection rate of 97.02% for a total of 1,210 data files (approximately 687 hours) which comprises 669 general videos files (about 424 hours) and 541 harmful video files (about 263 hours). It's speed is 5.6 times faster than the traditional methods whitout reducing the detection rate.

• Journal of KIISE
• /
• v.43 no.11
• /
• pp.1281-1297
• /
• 2016

In recent years, personal videos have seen a tremendous growth due to the substantial increase in the use of smart devices and networking services in which users create and share video content easily without many restrictions. However, taking both into account would significantly improve event detection performance because videos generally have multiple modalities and the frame data in video varies at different time points. This paper proposes an event detection method. In this method, high-level features are first extracted from multiple modalities in the videos, and the features are rearranged according to time sequence. Then the association of the modalities is learned by means of DNN to produce a personal video event detector. In our proposed method, audio and image data are first synchronized and then extracted. Then, the result is input into GoogLeNet as well as Multi-Layer Perceptron (MLP) to extract high-level features. The results are then re-arranged in time sequence, and every video is processed to extract one feature each for training by means of DNN.

• Journal of the Korean Society for information Management
• /
• v.31 no.3
• /
• pp.181-200
• /
• 2014

This study aims to analyze and evaluate video search services of major search portals, Naver and Google Korea. In particular, this study analyzed characteristics such as collection distribution, yearly distribution, the ratio of redundant search results, the ratio of advertising, and the quality of videos. This study also evaluated relevance, credibility, and currency of video search results, and investigated the factors that influence relevance and credibility. Finally, types and characteristics of error results were analyzed. The results of this study show that the relevance of Google's video search results is higher than those of Naver, whereas currency of Naver's search results is somewhat higher than those of Google. Google has more high resolution videos than Naver, and Naver has more advertising than Google. Both Google and Naver return many redundant videos in the search results. The results of this study can be implemented to the portal's effective development of video search services.

• Journal of the Korean Society for information Management
• /
• v.34 no.4
• /
• pp.227-245
• /
• 2017

This study mainly investigates the motivations of YouTube and Flicker users for posting videos or images/photos on each service. The motivational framework with ten factors such as enjoyment, self-efficacy, learning, personal gain, altruism, empathy, social engagement, community interest, reputation and reciprocity were used to test the motivations. Those who are users of YouTube and Flickr were recruited from Amazon Mechanical Turk to participate in online surveys. Findings show that learning and social engagement are the two most highly rated motivations. Altruism was rated relatively low, although it was strongly correlated with all other motivations. Personal gain was rated as the lowest by both users but Flickr users rated personal gain higher than YouTube users. Findings from this study could be applicable to specify user motivations for using the services and to upgrade the designs of the services in the future.

• Journal of KIISE:Computer Systems and Theory
• /
• v.35 no.3
• /
• pp.103-110
• /
• 2008

This paper proposes an effective keyword extraction method for the Web videos. The proposed method classifies the Web video pages in one of 4 types. As such, we analyzed the structure of the Web pages based on the number of videos and the layout of the Web pages. And then we applied the keyword extraction algorithm fit to each page type. The experiment with 1,087 Web pages that have total 2,462 videos showed that the recall of the proposed extraction method is 18% higher than ImagerRover[2]. So, the proposed method could be used to build a powerful video search system for WWW.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.06a
• /
• pp.619-622
• /
• 2007

PDA provides an opportunity for users to study anytime and anywhere because it is portable and convenient thanks to its relatively small size. However, users may face difficulties to fully recognize the characters provided through lecture videos, due to its low resolution and small scaled screen. This thesis proposes a system of remote lecture in which the size of videos can be adjusted and transmitted on the basis of contents necessary for study, using detection of region-of-interest(ROI) image, and a method of image scaling in a bid to solve such a problem of PDAs. The experiment on 802.11b wireless network shows that the proposed system is able to provide more optimized lecture videos than in existing method.

• Journal of Advanced Navigation Technology
• /
• v.18 no.3
• /
• pp.209-214
• /
• 2014

Current black boxes have a problem of storing unnecessary imagery data recordings without data classification. For this reason, users have to erase videos every time. This method is inadequate for black boxes with limited memory capacity. In this paper, we design and implement a system that recognizes traffic accident situations and saves these recordings by classifying them according to weighted values. The system was made to save video recorded at a 30-sec interval of every event to black box folders by changing names based on weighted value data under the external environment in a 1:10 scale model car. Based on this, when the tests were performed as a major car accident while driving, the videos were created in w2 folder, and when the tests were performed as a minor car accident while stopped, the videos were created in w1 folder.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.2
• /
• pp.76-84
• /
• 2014

As the number of high-density videos increase, parallel processing approaches are necessary to process a large-scale of video data. When a processing method of video data requires thousands of simple operations, GPU-based parallel processing is preferred to CPU-based parallel processing by way of reducing the time and space complexities of a given computation problem. This paper studies the parallel design and implementation of a shot-boundary detection algorithm. The proposed shot-boundary detection algorithm uses pixel brightness comparisons and global histogram data among the blocks of frames, and the computation of these data is characterized with the high parallelism for the related operations. In order to maximize these operations in parallel, the computations of the pixel brightness and histogram are designed in parallel and implemented in NVIDIA GPU. The GPU-based shot detection method is tested with 10 videos from the set of videos in National Archive of Korea. In experiments, the detection rate is similar but the computation time is about 10 time faster to that of the CPU-based algorithm.