• Journal of Digital Contents Society
• /
• v.18 no.4
• /
• pp.777-785
• /
• 2017

In the modern real world, we can extract and recognize some texts to get a lot of information from the scene containing them, so the techniques for extracting and recognizing text areas from a scene are constantly evolving. They can be largely divided into texture-based method, connected component method, and mixture of both. Texture-based method finds and extracts text based on the fact that text and others have different values such as image color and brightness. Connected component method is determined by using the geometrical properties after making similar pixels adjacent to each pixel to the connection element. In this paper, we propose a method to adaptively change to improve the accuracy of text region extraction, detect and correct the slope of the image using edge and image segmentation. The method only extracts the exact area containing the text by correcting the slope of the image, so that the extracting rate is 15% more accurate than MSER and 10% more accurate than EEMSER.

• Journal of Convergence for Information Technology
• /
• v.7 no.2
• /
• pp.31-36
• /
• 2017

In this paper, we try to search and reproduce the image part of a specific cast from a large number of images. The conventional method must manually set the offset value when searching for a scene or viewing a corner. However, in this paper, the proposed method learns the main character 's face, then finds the main character in the image recognition and moves to the scene where the main character appears to reproduce the image. Data for specific performers is extracted and collected using crawl techniques. Based on the collected data, we learn using convolutional neural network algorithm and perform performance evaluation using it. The performance evaluation measures the accuracy by extracting and judging a specific performer learned in the extracted key frame while playing the drama. The performance confirmation of how quickly and accurately the learned scene is searched has obtained about 93% accuracy. Based on the derived performance, it is applied to the image service such as viewing, searching for person and detailed information retrieval per corner

• Journal of Korea Game Society
• /
• v.17 no.3
• /
• pp.71-82
• /
• 2017

This study was conducted to understand the relationship between teacher's game recognition types and the acceptance of student game use in school. We surveyed 250 elementary/middle/high school teachers, school counselors and professional counselors. Four game recognition types have differences in gender, age, game experience, attitude toward game, response to game, and the evaluation of academic influence. When we compared the acceptance of student game use according to teacher's game recognition types, the differences in 'need for supervision', 'willingness to use', 'concern and monitoring' and 'acceptance as alternative activities' were identified. However, all types showed low scores of 'acceptance efficacy'. Finally, we discussed what kind of support would be effective to increase the acceptance of game use in school.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10c
• /
• pp.562-564
• /
• 1998

멀티미디어 기술이 발전됨에 따라 방대한 양의 동영상 데이터로부터 원하는 장면을 빠르고 손쉽게 검색하기 위한 연구는 동영상을 포함한 멀티미디어 서비스가 제공되는 현 시점에서 대단히 시급한 문제로 인식되고 있다. 본 연구에서는 동영상데이터를 대상으로 장면전환 지점을 검출하기 위하여 검색을 원하는 영상을 4개의 구역으로 구분하여 각각의 구역에 경계위치를 설정하며, 설정된 경계위치에 해당하는 칼라값을 분석하여 비디오 메모리에 저장한 후 이미지 데이터 베이스에 저장된 동영상 데이터의 각 Frame도 공간별 칼라값을 추출하여 이를 비디오 메모리값과 비교해서 연속된 두 Frame간의 칼라값 차이를 구하여 그 차이가 임계값 이상이 되면 장면전환 지점으로 검출하는 방법으로 사용자가 원하는 정보를 빠르게 검색하는 시스템을 설계하였다.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.5
• /
• pp.1130-1139
• /
• 1998

This paper presents an integrated solution for the content-based news video indexing and the retrieval. Currently, it is impossible to automatically index a general video, but we can index a specific structural video such as news videos. Our proposed model extracts automatically the key frames by using the structured knowledge of news and consists of the news item segmentation, caption recognition and search browser modules. We present above three modules in the following: the news event segmentation module recognizes an anchor-person shot based on face recognition, and then its news event are divided by the anchor-person's frame information. The caption recognition module detects the caption-frames with the caption characteristics, extracts their character region by the using split-merge method, and then recognizes characters with OCR software. Finally, the search browser module could make a various of searching mechanism possible.

• Proceedings of the Korean Information Science Society Conference
• /
• 2011.06c
• /
• pp.327-330
• /
• 2011

기계학습 기술이 발달함에 따라 기계학습은 제한된 상황에서 벗어나, 실생활과 비슷한 복잡하고 다양한 상황에서의 학습이 중요한 이슈가 되었다. 본고에서는 현실과 비슷한 상황을 도입하기 위하여 드라마를 사용한다. 드라마 내의 등장인물들은 말투, 어조, 관심주제와 같이 다양한 특성을 내재하고 있다. 등장인물들의 다양한 특성 중 관심주제는 대본 안에 글로 드러나 있으므로 기계학습을 통해 등장 인물의 인식에 활용할 수 있다. 최근, 확률그래프모델 분야에서 문서의 주제를 다루는 기법으로 자주 거론되는 토픽 모델 중 하나인 Author-Topic (AT) 모델은 등장인물의 관심주제를 학습하는 데에 적합하다. 본 논문에서는 AT 모델로 대본을 학습하고, 학습된 데이터 분포를 이용하여 장면에 등장하는 인물들을 인식하는 방법을 제시한다. 이 방법의 성능을 측정하기 위해, 미국 TV 드라마 'Friends' 대본 39편을 학습시키고, 장면에 대해 등장인물을 인식하는 실험을 수행하였다. 이 실험을 통해 본고에서 Author-Topic 모델을 이용한 인물 인식 방법이 다수의 인물이 참여한 담화의 인물들을 인식하는데 강점이 있음을 확인할 수 있다.

• Electronics and Telecommunications Trends
• /
• v.34 no.6
• /
• pp.133-144
• /
• 2019

Although deep learning-based visual image recognition technology has evolved rapidly, most of the commonly used methods focus solely on recognition accuracy. However, the demand for low latency and low power consuming image recognition with an acceptable accuracy is rising for practical applications in edge devices. For example, most Internet of Things (IoT) devices have a low computing power requiring more pragmatic use of these technologies; in addition, drones or smartphones have limited battery capacity again requiring practical applications that take this into consideration. Furthermore, some people do not prefer that central servers process their private images, as is required by high performance serverbased recognition technologies. To address these demands, the object and scene recognition technologies for mobile/embedded devices that enable optimized neural networks to operate in mobile and embedded environments are gaining attention. In this report, we briefly summarize the recent trends and issues of object and scene recognition technologies for mobile and embedded devices.

• Journal of KIISE:Software and Applications
• /
• v.34 no.6
• /
• pp.549-561
• /
• 2007

Vision-based scene understanding is to infer and interpret the context of a scene based on the evidences by analyzing the images. A probabilistic approach using Bayesian networks is actively researched, which is favorable for modeling and inferencing cause-and-effects. However, it is difficult to gather meaningful evidences sufficiently and design the model by human because the real situations are dynamic and uncertain. In this paper, we propose a learning method of Bayesian network that reduces the computational complexity and enhances the accuracy by searching an efficient BN structure in spite of insufficient evidences and training data. This method represents the domain knowledge as ontology and builds an efficient hierarchical BN structure under constraint rules that come from the ontology. To evaluate the proposed method, we have collected 90 images in nine types of circumstances. The result of experiments indicates that the proposed method shows good performance in the uncertain environment in spite of few evidences and it takes less time to learn.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.3
• /
• pp.384-390
• /
• 2020

The reverberation effect on the sound when producing movies or VR contents is a very important factor in the realism and liveliness. The reverberation time depending the space is recommended in a standard called RT60(Reverberation Time 60 dB). In this paper, we propose a scene recognition technique for automatic reverberation editing. To this end, we devised a classification model that independently trains color images and predicted depth images in the same model. Indoor scene classification is limited only by training color information because of the similarity of internal structure. Deep learning based depth information extraction technology is used to use spatial depth information. Based on RT60, 10 scene classes were constructed and model training and evaluation were conducted. Finally, the proposed SCR + DNet (Scene Classification for Reverb + Depth Net) classifier achieves higher performance than conventional CNN classifiers with 92.4% accuracy.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.659-662
• /
• 2008

In sports videos, text provides valuable information about the game such as scores and information about the players. This paper proposed a robust recognition method of player name in golf videos. In golf, most of users want to search the scenes which contain the play shots of favorite players. We use text information in golf videos for robust extraction of player information, By using OCR, we have obtained the text information, and then recognized the player information from player name DB. We can search the scenes of favorite players by using this player information. By conducting experiments on several golf videos, we demonstrate that our method achieves impressive performance with respect to the robustness.