• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2014.06a
• /
• pp.254-257
• /
• 2014

Recently, various applications using real-time face detection have been developed as face recognition technology and hardware grows. While network service is developing and video instruments costs lower, it is needed that smart surveillance camera and service using network camera based on IP and face detection technology. However, videos should be compressed for reducing network bandwidth and storage capacity in surveillance system. As it requires high-level improvement of system performance when all the compressed frames are processed in a face detection program, fast face detection method is needed. In this paper, not only a fast way of algorithm using Haar like features and adaboost learning and motion information but also an application on broadcast system is suggested.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10c
• /
• pp.307-312
• /
• 2006

멀티미디어의 비중이 커짐에 따라 컴퓨터 과학 각 분야에서 독자적인 기술들을 이용하여 실제 응용 및 시스템을 구축하고 있다. 하지만 멀티미디어 동영상 내에서 객체의 행위 단독적인 움직임을 수치로만 표현하여 자료를 처리함에 따라 의미를 해석하는 것이 부자연스럽고 정확한 숫자에 부합하는 행동의 검출이 어렵다. 본 논문에서는 멀티미디어 동영상의 기본적인 행위를 추출하고 이를 추상화, 정형화하여 보다 상위단계로 접근을 유도하여 멀티미디어 데이터에 대한 접근을 용이하게 하기위한 방법에 대하여 논의하였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1689-1694
• /
• 2005

Face detection plays an important role in face recognition, video surveillance, and human computer interface. In this paper, we present a face detection system using eye detection with progressive thresholding from a digital camera. The face candidate is detected by using skin color segmentation in the YCbCr color space. The face candidates are verified by detecting the eyes that is located by iterative thresholding and correlation coefficients. Preprocessing includes histogram equalization, log transformation, and gray-scale morphology for the emphasized eyes image. The distance of the eye candidate points generated by the progressive increasing threshold value is employed to extract the facial region. The process of the face detection is repeated by using the increasing threshold value. Experimental results show that more enhanced face detection in real time.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2015.11a
• /
• pp.165-168
• /
• 2015

최근 디지털 콘텐츠 및 멀티미디어는 거의 모든 IT 산업에서 중요한 역할을 하고 있으며, 이중에서도 비디오 콘텐츠는 그 동안의 검색, 유통, 관리 등의 수동적인 범위를 넘어 실세계와의 실시간 상호작용을 통한 능동적이며 혁신적인 역할을 수행하고 있다. 실시간 비디오 검색 기술은, 다양한 분야에서 활용이 가능한데, 대표적인 서비스 분야는 ALV(Autonomous Land Vehicle : 무인자동차), SNS 서비스, 오락/스포츠/광고 서비스, 모바일 쇼핑, AR, Surveillance 분야 등 매우 다양하다. 본 논문에서는, 실시간 비디오 검색 기술의 개요와 적용분야 및 사례를 설명하고 실시간 비디오 검색기술을 객관적으로 평가할 수 있는 방법, 절차에 대한 인증서에 대한 규격을 제정하여, 사용자가 표준화된 실시간 비디오 검색 기술의 인증서의 내용을 토대로 자신의 목적에 따라 기술을 선택하여 사용할 수 있도록 하였다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2014.11a
• /
• pp.56-58
• /
• 2014

본 논문에서는 압축 도메인에서 고속으로 움직임을 검출하고 해당 구간을 저장 하는 알고리즘을 제안한다. 제안하는 알고리즘은 H.264/AVC 기반의 압축 비트스트림에서 움직임 벡터와 참조프레임을 이용하여 움직임이 있는 프레임을 검출하고 움직임 유무에 따라 GOP 단위로 저장하는 과정을 수행한다. 압축도메인에서 움직임 검출과 구간 저장을 수행함으로써 복잡도를 낮추고 비디오 저장을 위한 공간을 절약해 실시간 다채널 영상 처리에 최적화 된 성능을 제공한다. 제안하는 움직임 검출 및 저장 시스템은 single thread 환경에서 실시간으로 평균 2957 프레임을 처리 가능하며, Multi thread의 경우 30 fps 영상 98개 채널을 실시간으로 처리 가능하다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.208-211
• /
• 2021

본 논문에서는 합성곱 신경망을 활용하여 영상에서 마스크 착용 및 미착용 상태를 탐지하는 방법을 제안한다. 코로나바이러스감염증-19(COVID-19)의 유행에 따라 감염 및 확산방지를 위해 마스크 정상적 착용이 요구되는데 몇몇 사람들은 이를 지키지 않고 있으며 현재의 감시 시스템은 입구에서 마스크 착용 여부를 검사하는 방식으로 작동될 뿐 공간에 입장한 다음 착용 여부를 알 수 없다. 제안하는 방법은 합성곱 신경망을 통해 영상에서 얼굴을 탐지하여 얻은 데이터를 이용하여 다수사람들의 마스크 착용 및 미착용 상태를 판별하는 방법으로 설계하였다.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.24 no.1
• /
• pp.97-106
• /
• 2014

CCTV(Closed Circuit television) is one of the widely used physical security technologies and video acquisition device installed at specific point with various purposes. Recently, as the CCTV capabilities improve, facial recognition from the information collected from CCTV video is under development. However, in case these technologies are exploited, concerns on major privacy infringement are high. Especially, a computer connected to a particular space images taken by the camera in real time over the Internet has emerged to show information services. In the privacy law, safety measures which is related with biometric template are notified. Accordingly, in this paper, for the protection of privacy video information in the video surveillance system, the technical and managerial requirements for video information security are suggested.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.7 no.1
• /
• pp.99-104
• /
• 2012

In this paper, implementation of a wireless video security system relates to a situation outside of using infrared sensors to detect changes when using Zigbee network security in the area of the sensor sends information to the server. The server can judge the situation if an emergency situation through the IP network security camera shot of the area to be transferred command to pantilte. The camera images and information in the security area, sent to administrator's smartphone users to control the camera can see the situation and More than a small video security system was designed so that user can monitor the security zone. Finally, for real-time to identify and respond to emergency situations based on the available wireless networks for video surveillance systems were verified through research and implementation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.751-770
• /
• 2019

Action recognition is an essential task in computer vision due to the variety of prospective applications, such as security surveillance, machine learning, and human-computer interaction. The availability of more video data than ever before and the lofty performance of deep convolutional neural networks also make it essential for action recognition in video. Unfortunately, limited crafted video features and the scarcity of benchmark datasets make it challenging to address the multi-person action recognition task in video data. In this work, we propose a deep convolutional neural network-based Effective Hybrid Learning (EHL) framework for two-person interaction classification in video data. Our approach exploits a pre-trained network model (the VGG16 from the University of Oxford Visual Geometry Group) and extends the Faster R-CNN (region-based convolutional neural network a state-of-the-art detector for image classification). We broaden a semi-supervised learning method combined with an active learning method to improve overall performance. Numerous types of two-person interactions exist in the real world, which makes this a challenging task. In our experiment, we consider a limited number of actions, such as hugging, fighting, linking arms, talking, and kidnapping in two environment such simple and complex. We show that our trained model with an active semi-supervised learning architecture gradually improves the performance. In a simple environment using an Intelligent Technology Laboratory (ITLab) dataset from Inha University, performance increased to 95.6% accuracy, and in a complex environment, performance reached 81% accuracy. Our method reduces data-labeling time, compared to supervised learning methods, for the ITLab dataset. We also conduct extensive experiment on Human Action Recognition benchmarks such as UT-Interaction dataset, HMDB51 dataset and obtain better performance than state-of-the-art approaches.

• Journal of the Institute of Convergence Signal Processing
• /
• v.7 no.3
• /
• pp.95-101
• /
• 2006

A face detection plays an important role in face recognition, video surveillance, and human computer interaction. In this paper, we propose a progressive threshold method to detect human faces in real time. The consecutive face images are acquired from camera and transformed into YCbCr color space images. The skin color of the input images are separated using a skin color filter in the YCbCr color space and some candidated face areas are decided by connected component analysis. The intensity equalization is performed to avoid the effect of many circumstances and an arbitrary threshold value is applied to get binary images. The eye area can be detected because the area is clearly distinguished from others in the binary image progressive threshold method searches for an optimal eye area by progressively increasing threshold from low values. After progressive thresholding, the eye area is normalized and verified by back propagation algorithm to finalize the face detection.