• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제6권1호
• /
• pp.15-22
• /
• 2017

인터넷 기술이 급속도로 발전함에 따라서 다양한 소셜 네트워크 서비스들이 나타나고 있다. 특히 스마트 기기들이 발전함에 따라서 소셜 네트워크 상에는 멀티미디어 콘텐츠가 넘쳐나고 있다. 그러나 불법적 목적을 가진 사용자에 의해 발생하는 범죄도 증가하면서 멀티미디어 포렌식을 이용한 콘텐츠 보호 및 불법 사용 차단의 필요성이 사회적으로 대두되고 있다. 본 논문에서는 멀티미디어 포렌식 기술의 하나로 디지털 동영상 획득 장치 판별을 위한 포렌식 기술을 제안한다. 먼저 모폴로지 필터링을 이용한 센서 패턴 노이즈 추출하는 방법을 제안한다. 이를 이용하여 참조 장치에서 촬영한 참조 동영상의 센서 패턴 노이즈를 추정하고, 검사 동영상에서 센서 패턴 노이즈를 추출한다. 그리고 두 센서 패턴 노이즈 사이의 유사성 계산을 통하여 검사 동영상이 참조 장치로 촬영을 했는지 판별을 수행한다. 제안한 기술의 성능 분석을 위하여 DSLR 카메라, 콤팩트 카메라, 캠코더, 액션캠 및 스마트폰 등을 포함한 총 30대의 장치에 대하여 개발한 알고리즘에 대한 정량적 성능 분석을 수행하였고, 그 결과 96%의 판별 정확도를 달성하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권6호
• /
• pp.2483-2503
• /
• 2016

Traffic congestion is a severe problem in many modern cities around the world. Real-time and accurate traffic congestion identification can provide the advanced traffic management systems with a reliable basis to take measurements. The most used data sources for traffic congestion are loop detector, GPS data, and video surveillance. Video based traffic monitoring systems have gained much attention due to their enormous advantages, such as low cost, flexibility to redesign the system and providing a rich information source for human understanding. In general, most existing video based systems for monitoring road traffic rely on stationary cameras and multiple vehicle tracking method. However, most commonly used multiple vehicle tracking methods are lack of effective track initiation schemes. Based on the motion of the vehicle usually obeys constant velocity model, a novel vehicle recognition method is proposed. The state of recognized vehicle is sent to the GM-PHD filter as birth target. In this way, we relieve the insensitive of GM-PHD filter for new entering vehicle. Combining with the advanced vehicle detection and data association techniques, this multiple vehicle tracking method is used to identify traffic congestion. It can be implemented in real-time with high accuracy and robustness. The advantages of our proposed method are validated on four real traffic data.

• 한국멀티미디어학회논문지
• /
• 제7권9호
• /
• pp.1263-1272
• /
• 2004

본 논문에서는 불법으로 복제된 컨텐츠의 근원지를 확인할 수 있는 핑거프린팅 기법을 제안하고 있다. 판매자와 구매자의 키로 만들어진 균일 랜덤 신호를 시간축 웨이블린 변환에 의해 얻어진 계수 중에서 배포 받을 사용자의 영역에 삽입하여 핑거프린팅을 수행하게 된다. 제안기법은 핑거프린팅된 컨텐츠에 대한 공모 공격과 MPEG2 합축에도 유일한 핑거프린팅 정보를 감지할 수 있다. 특히, 핑거프린팅 정보를 삽입할 사용자의 영역 지정을 위해서 시간축 웨이블릿 변환의 특성을 이용한다. 실험에서는 비디오 컨텐츠의 불법 배포를 추적할 수 있음을 보이고, 다양한 공모공격과 MPEG2 압축에 대해 강인(robustness)함을 보인다.

• 한국소음진동공학회논문집
• /
• 제21권11호
• /
• pp.1074-1083
• /
• 2011

Noise of digital camera has been noticeable to its users. Particularly, noise of a barrel assembly module in zoom in/zoom out operation is recorded while taking a video. Reduction of barrel noise becomes crucial but there are not many studies on noise of digital camera due to its short history of use. In this study, experiment-based analyses are implemented to identify sources of noise and vibration because of complexity and compactness of the barrel system. Output noise is acquired in various operation conditions using synchronization for spectral analysis. Noise sources of a barrel assembly in zoom operating are first identified by the comparison with gear frequency analysis and then correlation analysis between noise and vibration is applied to confirm the generation path of noise. Analysis on noise transfer characteristic of zoom module is also carried out in order to identify the most contributing components. One of possible countermeasures of noise in zoom operating is investigated by an experimental approach.

• Journal of Information Processing Systems
• /
• 제16권1호
• /
• pp.6-29
• /
• 2020

Biometrics identification using multiple modalities has attracted the attention of many researchers as it produces more robust and trustworthy results than single modality biometrics. In this paper, we present a novel multimodal recognition system that trains a deep learning network to automatically learn features after extracting multiple biometric modalities from a single data source, i.e., facial video clips. Utilizing different modalities, i.e., left ear, left profile face, frontal face, right profile face, and right ear, present in the facial video clips, we train supervised denoising auto-encoders to automatically extract robust and non-redundant features. The automatically learned features are then used to train modality specific sparse classifiers to perform the multimodal recognition. Moreover, the proposed technique has proven robust when some of the above modalities were missing during the testing. The proposed system has three main components that are responsible for detection, which consists of modality specific detectors to automatically detect images of different modalities present in facial video clips; feature selection, which uses supervised denoising sparse auto-encoders network to capture discriminative representations that are robust to the illumination and pose variations; and classification, which consists of a set of modality specific sparse representation classifiers for unimodal recognition, followed by score level fusion of the recognition results of the available modalities. Experiments conducted on the constrained facial video dataset (WVU) and the unconstrained facial video dataset (HONDA/UCSD), resulted in a 99.17% and 97.14% Rank-1 recognition rates, respectively. The multimodal recognition accuracy demonstrates the superiority and robustness of the proposed approach irrespective of the illumination, non-planar movement, and pose variations present in the video clips even in the situation of missing modalities.

• 정보보호학회논문지
• /
• 제26권4호
• /
• pp.885-894
• /
• 2016

스마트폰이 널리 보급됨에 따라 누구나 쉽게 사진과 동영상을 촬영하고 배포할 수 있는 시대가 되었다. 개인이 스마트폰으로 촬영한 동영상은 중요한 수사 단서나 증거로 활용되고 이때 동영상이 특정 스마트폰으로 촬영되었음을 입증해야 하는 상황이 발생한다. 이를 위해 기존 연구들에서 제시한 다양한 방식의 fingerprint 기법을 활용할 수 있다. 하지만 fingerprint 기법을 사용한 결과의 신빙성을 보강해야 하거나 그 기법을 활용할 수 없는 상황들이 존재한다. 따라서 fingerprint 기법의 사용 이전에 스마트폰 포렌식 조사가 선행되어야 하고, 동영상 파일의 메타데이터 정보를 정리한 데이터베이스를 구축할 필요가 있다. 본 논문에서는 동영상 촬영이 스마트폰에 남기는 아티팩트와 상기한 데이터베이스에 대해 설명하고자 한다.

• E2M - 전기 전자와 첨단 소재
• /
• 제8권5호
• /
• pp.633-639
• /
• 1995

Dropouts in magnetic media are a primary quality deficiency which is detrimental to magnetic signal quality and thus the major impediment to error-free recording, especially in high density digital recorders. The specific form of defects and causes found to be responsible for signal dropouts occurring in magnetic tape were studied. Dropout occurred when the RF signal falls to low level due to the various types of surface defects. However, the fall in the level of the RF signal in gravure roll coated tapes was most often caused by foreign particles adhering to the magnetic tape. It was also shown from the present study that scattered particles trapped under the tape surface or put on the top can lift it as it crosses the head, creating a spherical tent shaped defect and causing a temporary signal drop. Dropout producing substances were identified through optical and microchemical analyses. The results were correlated with measured electrical signal losses combined with analytical microscopy to clarify source identification of defects.