• Journal of Broadcast Engineering
• /
• v.12 no.4
• /
• pp.311-320
• /
• 2007

In this paper, an efficient detection algorithm of an anchor shot in news video is presented. We observed the audio visual characteristics of news video and proposed several low level features which are appropriate for detecting an anchor shot in news video. The overall structure of the proposed algorithm is composed of 3 stages: the pause detection, the audio cluster classification, and the matching with motion activity stage. We used the audio features as well as the motion feature in order to improve the indexing accuracy and the simulation results show that the performance of the proposed algorithm is quite satisfactory.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.07a
• /
• pp.661-662
• /
• 2020

현대사회의 산업 현장에서 작업효율과 안전사고예방은 기업의 이익과 직결된다. 현장에서의 인력의 사용으로 인한 한계점을 가지고 있기 때문에 효율적이고 안정적으로 작업 효율을 내며 현장의 많은 안전사고를 미연에 방지하기 위해 많은 산업현장들은 4차 산업 혁명을 통해 수많은 작업들을 로봇을 이용한 자동화로 대체해 오고 있다. 단순히 짐을 옮기고 재고를 파악할 뿐인 간단한 작업임에도 불구하고 물류 피킹/분류 작업은 아직까지 인력을 사용한다. 인력을 한계를 극복하기 위해 작업 현장을 라인 트레이서를 통해 이동하고, 영상분석을 이용해 로봇 암으로 원하는 물건을 정확하게 피킹하고자 적재 하도록 설계한 '영상분석을 이용한 자동 피킹.분류시스템' 기술을 제안한다. 기존의 단순 반복 노동의 피킹/분류 작업을 수행하며 영상분석을 통해 어플리케이션을 이용하여 재고 관리또한 가능하다,

• Journal of KIISE:Software and Applications
• /
• v.32 no.12
• /
• pp.1247-1259
• /
• 2005

This paper proposes a real-time face recognition system which detects multiple faces from low resolution video such as web-camera video. Face recognition system consists of the face detection step and the face classification step. At First, it finds face region candidates by using AdaBoost based object detection method which have fast speed and robust performance. It generates reduced feature vector for each face region candidate by using principle component analysis. At Second, Face classification used Principle Component Analysis and multi-SVM. Experimental result shows that the proposed method achieves real-time face detection and face recognition from low resolution video. Additionally, We implement the auto-tracking face recognition system using the Pan-Tilt Web-camera and radio On/Off digital door-lock system with face recognition system.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2007.11a
• /
• pp.279-283
• /
• 2007

움직이는 물체를 분류하는 것은 영상 감시 시스템에서 가장 중요한 분야 중의 하나이다. 사람과 자동차는 영상 감사 시스템에서 인식해야 하는 가장 중요한 물체의 종류이기 때문에 본 연구에서는 인식하는 물체의 종류를 이것들로 제한한다. 사용되는 특성으로는 물체의 움직임에서 추출되는 특성과 형태에서 추출되는 특성이 있다. 이 두 가지 특성들은 정지된 하나의 카메라로부터 입력된 영상에 나타나는 물체를 분류하기 위하여 사용된다. 움직임으로부터 추출되는 특성은 연결 성분 분석을 이용한 물체 추적과 밀접한 관련이 있다. 그리고 형태 기반 특성에 관한 학습은 종횡비(aspect ratio)와 4개의 윤곽선을 가지고 수행된다. 움직임 기반 특성과 종횡비는 물체를 사람과 자동차로 구분하는데 이용되고 각각의 종류를 더욱 세분화하기 위하여 4개의 윤곽선이 사용된다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2001.11b
• /
• pp.149-154
• /
• 2001

멀티미디어 정보표현 기술인 MPEG-7 표준이 빠른 속도의 진전을 보임에 따라 이를 활용한 검색 기술 개발도 활발히 진행 중에 있다 방대한 량의 동영상 내용 검색 기술 연구에 있어서 우선적으로 고려되어야 할 부분이 내용이 연속되는 프레임들의 분류이다. 이를 위해서는 물리적인 장면전환이 이루어지는 부분에 대한 실시간 자동 cut detection 기술 및 이 컷 프레임 영상에 대한 내용 기술을 자동적으로 수행할 필요성이 있다. 각 컷 프레임의 자동 내용 기술의 전처리로써 본 논문에서는 장면전환이 생기는 프레임의 영상의 어떠한 정보도 사전 정보로 취하지 않고 사용자의 개입이 없는 상황에서 영상의 컬러 특성 및 에지 정보만을 가지고 shape 영역 segmentation을 자동으로 실행하는 방법을 제안한다. 제안한 방법의 성능은 segmentation된 영상과 원 영상과의 영역비교를 통한 유사도에 의해 평가하며, 시뮬레이션 결과에서 제안한 알고리즘은 평균 90%이상의 영역 분할이 정확하게 됨을 알 수 있었고, 컬러의 구분이 명확하지 않은 자연영상에서도 robust한 segmentation 결과를 나타냄을 본 연구를 통하여 알 수 있었다.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.1
• /
• pp.588-594
• /
• 2021

Due to the recent proliferation of video contents, previous contents expressed as the character or the picture are being replaced to video and growth of video contents is being boosted because of emerging new platforms. As this accelerated growth has a great impact on the process of universalization of technology for ordinary people, video production and editing technologies that were classified as expert's areas can be easily accessed and used from ordinary people. Due to the development of these technologies, tasks like that recording and adjusting that depends on human's manual involvement could be automated through object tracking technology. Also, the process for situating the object in the center of the screen after finding the object to record could have been automated. Because the task of setting the object to be tracked is still remaining as human's responsibility, the delay or mistake can be made in the process of setting the object which has to be tracked through a human. Therefore, we propose a novel object tracking technique of CMT combining the face detection technique utilizing Haar cascade classifier. The proposed system can be applied to an effective and robust image tracking system for continuous object tracking on the smartphone in real time.

• The Journal of the Korea Contents Association
• /
• v.12 no.12
• /
• pp.1-13
• /
• 2012

3D video is regarded as the next generation contents in numerous applications. The 2D-to-3D video conversion technologies are strongly required to resolve a lack of 3D videos during the period of transition to the full ripe 3D video era. In 2D-to-3D conversion methods, after the depth image of each scene in 2D video is estimated, stereoscopic video is synthesized using DIBR (Depth Image Based Rendering) technologies. This paper proposes a novel depth fusion algorithm that integrates multiple depth cues contained in 2D video to generate stereoscopic video. For the proper depth fusion, it is checked whether some cues are reliable or not in current scene. Based on the result of the reliability tests, current scene is classified into one of 4 scene types and scene-adaptive depth fusion is applied to combine those reliable depth cues to generate the final depth information. Simulation results show that each depth cue is reasonably utilized according to scene types and final depth is generated by cues which can effectively represent the current scene.

• Journal of Korea Multimedia Society
• /
• v.12 no.7
• /
• pp.913-921
• /
• 2009

Quantization is an essential method for compression of video. The quantizer can adjust the bitrate and control the picture quality. Especially, the subjective picture quality can be improved if the concerned region within a video sequence has good picture quality. In this paper, firstly a classification method according to the subjective concerned region within the video sequence is suggested. Also we propose a method that assigns the quantization step-size differentially according to the concerned region within the video. Totally subjective picture quality can be increased by appling the quantization step-size as small value relatively for the concerned region compared with the other regions. We can find the result that the proposed method gives the improved picture quality by assigning differently quantization step-size and the best improvement can be brought when the difference between maximum and minimum values of the quantization step-size in a picture is from 4 to 8.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.3
• /
• pp.319-327
• /
• 2012

In this paper, we present a sign language recognition model which does not use any wearable devices for object tracking. The system design issues and implementation issues such as data representation, feature extraction and pattern classification methods are discussed. The proposed data representation method for sign language patterns is robust for spatio-temporal variances of feature points. We present a feature extraction technique which can improve the computation speed by reducing the amount of feature data. A neural network model which is capable of incremental learning is described and the behaviors and learning algorithm of the model are introduced. We have defined a measure which reflects the relevance between the feature values and the pattern classes. The measure makes it possible to select more effective features without any degradation of performance. Through the experiments using six types of sign language patterns, the proposed model is evaluated empirically.

• Journal of Broadcast Engineering
• /
• v.23 no.6
• /
• pp.886-895
• /
• 2018

In this paper, we introduce a technique to detect the video forgery by using the regularity that occurs in the video compression process. The proposed method uses the hierarchical regularity lost by the video double compression and the frame deletion. In order to extract such irregularities, the depth information of CU and TU, which are basic units of HEVC, is used. For improving performance, we make a depth map of CU and TU using local information, and then create input data by grouping them in GoP units. We made a decision whether or not the video is double-compressed and forged by using a general three-dimensional convolutional neural network. Experimental results show that it is more effective to detect whether or not the video is forged compared with the results using the existing machine learning algorithm.