• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.06a
• /
• pp.123-124
• /
• 2018

본 논문에서는 HEVC 국제표준으로 압축된 비디오에서 움직임 물체를 추적하는 방법을 제안한다. 제안하는 방법은 HEVC 비트스트림의 대부분을 차지하는 인터 프레임의 움직임 벡터 정보를 입력 데이터로 사용하는 ST-MRF(Spatio-Temporal-Markov Random Field) 모델을 기반으로 하며, ST-MRF 모델에서 발생할 수 있는 오차전파로 점진적으로 객체를 부정확하게 추적하는 것을 HEVC GOP(Group of Picture)마다 삽입되는 인트라 프레임만을 복호화 하여 픽셀 정보를 입력으로 하는 YOLO 모델과 융합시켜 보정함으로써 추적하던 객체를 잃지 않고 강건하게 추적하는 방법을 제안한다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.26 no.4
• /
• pp.527-537
• /
• 1999

비디오 데이터를 이용한 응용 프로그램을 개발하기위해서 비디오 데이터베이슬ㄹ 구축하고자하는 경우에는 비디오의 내용(Content)에 따라 자동으로 장면 변화를 검출(Scene Change Detection)하는 기술이 필요하다. 본 논문에서는 MPEG-1 형식으로 저장된 비디오 데이터에 대하여 장면의 변화를 자동적으로 검출할 수 있는 방법을 제안하고 실험을 통하여 그 유용성을 보인다. 제안한 검출 방법에서는 B 프레임의 각 매크로 블록들에 대하여 시간적으로 과거 B 프레임의 대응되는 매크로블록의 타입과 비교를 하고, 이런 각 매크로블록들에 대한 비교 결과의 합이 입계치보다 큰 경우에 장면이 변한 것으로 판단한다. 제안한 방법에서는 입력 비디오 스트림에서 B 프레임의 매크로블록층 정보만을 이용하여 I프레임과 P 프레임의 장면 변화 검출도 가능하므로 정교한 검출이 가능하다. 또한 이런 검출 방법은 단순히 한 B 프레임안의 매크로 블록개수만을 조사하여 장면 변화여부를 검출하는 기존의 방법에 대하여 각 매크로블록의 타입 정보뿐만 아니라 위치 정보도 이용하기 때문에 장면 변화 검출이 견고하다. MPEG-1 형식으로 부호화한 뉴스 및 영화 비디오 데이터에 대한 실험에 의하면, 본 논문에서 제안한 검출 방법은 95% 이상의 정확성을 보임을 알 수있다. 본 논문에서 제안한 MPEG-1 비디오 장면변화검출방법은 MPEG-1 형식의 비디오 데이터를 이용한 디지털 라이브러리 등의 구축등에 유용하게 사용될수 있을 것이다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.9 no.5
• /
• pp.943-950
• /
• 2005

Recently, many of researches on stream transmission for satisfying each of different real-time transmission condition of the multimedia data that demands various service quality through high-speed networks have been studied actively. In this paper, we design a scheme that discriminately reserves the network resources for the transmission of each multimedia application and propose a bandwidth allocation scheme for improving the utilization ratio of free resources. And we also propose a pipelining scheme for providing flexible real-time transmission. The proposed schemes can be used to support a real-time transmission by applying feedback transmission control method based on receiving buffer for guaranteeing the synchronization conditions requested by the multimedia data. Moreover, we propose a transmission control scheme that can take the amount of network resources down to the minimum amount within the range of permissible error-range under the guarantee with no quality degradation simultaneously when the bottleneck is caused by the network congestion. Finally, we propose a dynamic bandwidth smoothing scheme that can smooth the maximum bandwidth to the demand of each video steam for giving continuous transmission to the delay sensitive video steam between senders and receivers.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.1
• /
• pp.370-381
• /
• 2012

Multi-criteria result extraction is crucial in many scientific applications that support real-time stream processing, such as habitat research and disaster monitoring. Skyline evaluation is computational intensive especially over continuous time-interval data streams where each object has its own customized expiration time. In this work, we propose TI-Sky - a continuous skyline evaluation framework. To ensure correctness, the result space needs to be continuously maintained as new objects arrive and older objects expire. TI-Sky strikes a perfect balance between the costs of continuously maintaining the result space and the costs of computing the final skyline result from this space whenever a pull-based user query is received. Our key principle is to incrementally maintain a partially precomputed skyline result space - however doing so efficiently by working at a higher level of abstraction. TI-Sky's algorithms for insertion, deletion, purging and result retrieval exploit both layers of granularity. Our experimental study demonstrates the superiority of TI-Sky over existing techniques to handle a wide variety of data sets.

• Journal of the Institute of Convergence Signal Processing
• /
• v.14 no.1
• /
• pp.62-69
• /
• 2013

This paper proposes real-time video transmission system by the car-mounted cameras based on MOST Network. Existing vehicles transmit videos by connecting the car-mounted cameras in the form of analog. However, the increase in the number of car-mounted cameras leads to development of the network to connect the cameras. In this paper, DSP is applied to process MPEG 2 encoding/decoding for real-time video transmission in a short period of time. MediaLB is employed to transfer data stream between DSP and MOST network controller. During this procedure, DSP cannot transport data stream directly from MediaLB. Therefore, FPGA is used to deliver data stream transmitting MediaLB to DSP. MediaLB is designed to streamline hardware/software application development for MOST Network and to support all MOST Network data transportation methods. As seen in this paper, the test results verify that real-time video transmission using proposed system operates in a normal matter.

• The KIPS Transactions:PartD
• /
• v.18D no.4
• /
• pp.215-224
• /
• 2011

U-Health collects patient information with various kinds of sensor. Stream data can be summarized as an interval event which has aninterval between start-time-point and end-time-point. Most of temporal mining techniques consider only the event occurrence-time-point and ignore stream data change of state. In this paper, we propose the temporal mining technique considering the event change of state in U-Health. Our method overcomes the restrictions of the environment by sending a significant event in U-Health from sensors to a server. We define four event states of stream data and perform the temporal data mining considered the event change of state. Finally, we can remove an ambiguity of discovered rules by describing cause-and-effect relations among events in temporal relation sequences.

• The KIPS Transactions:PartD
• /
• v.17D no.6
• /
• pp.453-462
• /
• 2010

The switch from analog to digital broadcast television is extended rapidly. The DTV can offer multiple programming choices, interactive capabilities and so on. Moreover, with the spread of Internet, the information exchange between the communities is increasing, too. These facts lead to the new TV service environment which can offer customized TV programs to personal/community users. This paper proposes a 'Customized Digital TV System for Individuals/Communities based on Data Stream Mining' which can analyze user's pattern of TV watching behavior. Due to the characteristics of TV program data stream and EPG(electronic program guide), the data stream mining methods are employed in the proposed system. When a user is watching DTV, the proposed system can control the surrounding circumstances as using the user behavior profiles. Furthermore, the channel recommendation system on the smart phone environment is proposed to utilize the profiles widely.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.11a
• /
• pp.1179-1180
• /
• 2011

유비쿼터스 컴퓨팅 환경에서 무한의 연속적으로 전송되는 데이터에 대한 처리가 요구되고 있다. 본 논문에서는 연속적이고 빠르게 발생하는 스트림 데이터로부터 유용한 정보를 발견하기 위한 기반 연구로써 트리거를 이용한 슬라이딩 윈도우 기반의 XML 빈발 구조 추출 방법을 제안한다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.2
• /
• pp.181-186
• /
• 2012

Recently there has been many active researches on searching similar sequences from data generated with the passage of time. Those data are classified as time series data or sequence data and have different semantics from scalar data of traditional databases. In this paper similar sequence search retrieves sequences that have a similar trend of value changes. At first we have transformed the original sequences by applying DFT. The converted data are more suitable for trend analysis and they require less number of attributes for sequence comparisons. In addition we have developed a region-based query and we applied bitmap indexes which could show better performance in data warehouse. We have built bitmap indexes with varying number of attributes and we have found the least cost query plans for efficient similar sequence searches.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.5
• /
• pp.1117-1132
• /
• 2019

Recently, IT technology such as internet, mobile, and IOT has rapidly developed, making it easy to collect data necessary for business, and the collected data is analyzed as a new method of big data analysis and used appropriately for business. In this way, data collection and analysis becomes easy. In such data, personal information including an identifier such as a sensor id, a device number, IP address, or the like may be collected. However, if systematic management is not accompanied by collecting and disposing of large-scale data, violation of relevant laws such as "Personal Data Protection Act". Furthermore, data quality problems can also occur and make incorrect decisions. In this paper, we propose a new data governance maturity model(DGMM) that can identify the personal data contained in the data collected by companies, use it appropriately for the business, protect it, and secure quality. And we also propose a over all implementation process for DG Program.