• 한국정보통신학회논문지
• /
• 제9권5호
• /
• pp.1019-1024
• /
• 2005

MPEG-2로 부호화 된 영상에서 발생하는 전송 오류는 화질의 열화를 가져오고, 시공간적으로 오류를 전파시킨다. 특히 비디오 비트열에서 헤더 정보의 오류는 복호화 과정 전체에 영향을 미치므로 데이터 정보의 오류와 달리 전체 영상에 심각한 화질의 열화를 일으킬 수 있다. 따라서 헤더 정보에서의 오류를 복원하는 것은 데이터 정보에서 오류를 복원하는 것보다 더 중요하다. 본 논문에서는 LSB(least significant bit) 부호화를 이용하여 헤더 정보를 양자화 된 DCT(discrete cosine transform) 계수에 반복적으로 삽입하여 전송함으로써 MPEG-2의 신택스 구조 그대로 유지하면서 헤더 정보의 오류를 복원할 수 있는 방법을 제안한다.

• IEIE Transactions on Smart Processing and Computing
• /
• 제4권6호
• /
• pp.422-433
• /
• 2015

In this paper, we demonstrate inter-layer prediction tools for scalable video coders. The proposed scalable coder is designed to support not only spatial, quality and temporal scalabilities, but also view scalability. In addition, we propose quad-tree inter-layer prediction tools to improve coding efficiency at enhancement layers. The proposed inter-layer prediction tools generate texture prediction signal with exploiting texture, syntaxes, and residual information from a reference layer. Furthermore, the tools can be used with inter and intra prediction blocks within a large coding unit. The proposed framework guarantees the rate distortion performance for a base layer because it does not have any compulsion such as constraint intra prediction. According to experiments, the framework supports the spatial scalable functionality with about 18.6%, 18.5% and 25.2% overhead bits against to the single layer coding. The proposed inter-layer prediction tool in multi-loop decoding design framework enables to achieve coding gains of 14.0%, 5.1%, and 12.1% in BD-Bitrate at the enhancement layer, compared to a single layer HEVC for all-intra, low-delay, and random access cases, respectively. For the single-loop decoding design, the proposed quad-tree inter-layer prediction can achieve 14.0%, 3.7%, and 9.8% bit saving.

• ETRI Journal
• /
• 제44권3호
• /
• pp.476-490
• /
• 2022

With the recent advances in technology, automatic speech recognition (ASR) has been widely used in real-world applications. The efficiency of converting large amounts of speech into text accurately with limited resources has become more vital than ever. In this study, we propose a method to rapidly recognize a large speech database via a transformer-based end-to-end model. Transformers have improved the state-of-the-art performance in many fields. However, they are not easy to use for long sequences. In this study, various techniques to accelerate the recognition of real-world speeches are proposed and tested, including decoding via multiple-utterance-batched beam search, detecting end of speech based on a connectionist temporal classification (CTC), restricting the CTC-prefix score, and splitting long speeches into short segments. Experiments are conducted with the Librispeech dataset and the real-world Korean ASR tasks to verify the proposed methods. From the experiments, the proposed system can convert 8 h of speeches spoken at real-world meetings into text in less than 3 min with a 10.73% character error rate, which is 27.1% relatively lower than that of conventional systems.

• 한국정보과학회논문지:정보통신
• /
• 제36권5호
• /
• pp.425-436
• /
• 2009

최근 광 대역 통합 네트워크에서 고품질의 멀티미디어 서비스에 대한 사용자 요구가 증가하고 있다. 또한, 사용자 단말기기의 다양화 및 대화면 디스플레이 장치의 보급으로 다양한 형태의 서비스 품질(QoS)에 대한 요구도 증가하고 있다. 이를 위해 네트워크 관점에서 동적 큐 관리 알고리즘과 같은 인터넷 성능을 개선하여 서비스 품질을 보장하는 연구와 종단 관점에서 미디어의 품질을 보장하기 위한 SVC(Scalable Video Coding) 부호화 기법에 대한 연구가 활발히 진행 중에 있다. 그러나, 기존 동적 큐 관리 알고리즘은 비디오 부호화 기술의 본질적인 특성에 대하여 고려하지 못하여 서비스 품질을 보장하는 못하는 문제점을 가지고 있다. 본 논문에서는 현재 혼잡제어 알고리즘의 문제점을 개선하기 위해 NAL (Network Abstract Layer)의 헤더 내 TID (Temporal_ID)를 통해 SVC 부호화 기술의 특성을 파악하여 프레임간 의존성이 낮은 프레임의 패킷에 대하여 차등적으로 패킷을 폐기하는 75-AQM (Temporal Scalability - Active Queue Management) 알고리즘을 제안하였다. 제안한 75-AQM 알고리즘은 혼잡상황 시 차등적인 패킷 폐기를 통해 SVC 부호화 기법을 이용하는 스트리밍 서비스에 대하여 안정적인 비디오 복호화를 통해 멀티미디어 서비스 품질을 보장하였다.

• Journal of information and communication convergence engineering
• /
• 제1권4호
• /
• pp.177-182
• /
• 2003

Conventional multilayer feedforward artificial neural networks are very effective in dealing with spatial problems. To deal with problems with time dependency, some kinds of memory have to be built in the processing algorithm. In this paper we show how the newly proposed Serial Input Neuron (SIN) convolutional decoders can be derived. As an example, we derive the SIN decoder for rate code with constraint length 3. The SIN is tested in Gaussian channel and the results are compared to the results of the optimal Viterbi decoder. A SIN approach to decode convolutional codes is presented. No supervision is required. The decoder lends itself to pleasing implementations in hardware and processing codes with high speed in a time. However, the speed of the current circuits may set limits to the codes used. With increasing speeds of the circuits in the future, the proposed technique may become a tempting choice for decoding convolutional coding with long constraint lengths.

• International Journal of Computer Science & Network Security
• /
• 제23권11호
• /
• pp.73-76
• /
• 2023

We introduce DCNN and DRAE appraoches for compression of medical videos, in order to decrease file size and storage requirements, there is an increasing need for medical video compression nowadays. Using a lossy compression technique, a higher compression ratio can be attained, but information will be lost and possible diagnostic mistakes may follow. The requirement to store medical video in lossless format results from this. The aim of utilizing a lossless compression tool is to maximize compression because the traditional lossless compression technique yields a poor compression ratio. The temporal and spatial redundancy seen in video sequences can be successfully utilized by the proposed DCNN and DRAE encoding. This paper describes the lossless encoding mode and shows how a compression ratio greater than 2 (2:1) can be achieved.

• 한국콘텐츠학회논문지
• /
• 제6권2호
• /
• pp.65-74
• /
• 2006

MPEG-2 비디오 압축열은 복잡한 부호화 알고리즘을 이용하여 압축하기 때문에 전송 오류에 매우 민감하다. 만약 패킷을 잃어버리거나 수신된 패킷에 오류가 있으면 현재 화면에 화질저하가 발생할 뿐만 아니라 화면수가 제한적이긴 하지만 뒤이어서 재생되는 화면에도 오류가 전파된다. 따라서 이런 전송오류의 영향을 막거나 최소화 하기위해서 다양한 오류 강인 부호화/복호화를 적용한다. 대표적인 오류 강인 방법이 오류 은폐 기법이다. 오류 은폐 기법은 손상된 비디오 데이터를 은폐하기 위해서 정상적으로 수신된 데이터의 공간적, 시간적 중복성을 이용한다. 손상된 데이터를 복원하기 위해 움직임 벡터를 추정하고 움직임 보상하는 것은 좋은 방법이다. 이 논문에서는 다양한 움직임 벡터 복원 방법에 기반한 오류 은폐 기법을 제안하고 일반적인 방법들과 성능을 비교한다.

• 한국발생생물학회지:발생과생식
• /
• 제16권1호
• /
• pp.69-75
• /
• 2012

Hatching occurred in the time dependent manners and strictly controlled. Although, the hatching processes are under the control of muti-embryotrophic factors and the expressed G proteins of cell generate integrated activation, the knowledge which GPCRs are expressed during hatching stage embryos are very limited. In the present study, which G proteins are involved was examined during blastocyst development to the hatching stage. The early-, expanded-, and lobe-stage blastocysts were treated with various $G_{\alpha}$ activators and H series inhibitors, and examined developmental patterns. Pertusis toxin (PTX) improved the hatching rate of the early-stage blastocyst and lobe-formed embryos. Cholera toxin (CTX) suppressed the hatching of the early-stage blastocyst and expanded embryos. The effects of toxins on hatching and embryo development were changed by the H7 and H8. These results mean that PTX mediated GPCRs activation is signaling generator in the nick or pore formation in the ZP. In addition, PTX mediated GPCR activation induces the locomotion of trophectoderm for the escaping. CTX mediate GPCRs activation is the cause of suppression of hatching processes. Based on these data, it is suggested that various GPCRs are expressed in the periimplantation stage embryos and the integration of the multiple signals decoding of various signals in a spatial and temporal manner regulate the hatching process.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -3
• /
• pp.1527-1530
• /
• 2002

Video summary is one of the tools which can provide the fast and effective browsing fur a lengthy video. Video summary consists of many key-frames that could be defined differently depending on the video genre it belongs to. Consequently, the video summary constructed by the uniform manner might lead into inadequate result. Therefore, identifying the video genre is the important first step in generating the meaningful video summary. We propose a new method that can classify the genre of the video data in MPEG compressed bit-stream domain. Since the proposed method operates directly on the com- pressed bit-stream without decoding the frame, it has merits such as simple calculation and short processing time. In the proposed method, only the visual information is utilized through the spatial-temporal analysis to classify the video genre. Experiments are done for 6 genres of video: Cartoon, Commercial, Music Video, News, Sports, and Talk Show. Experimental result shows more than 90% of accuracy in genre classification for the well-structured video data such as Talk Show and Sports.

• 방송공학회논문지
• /
• 제13권1호
• /
• pp.119-127
• /
• 2008

Transmission of video over Internet or wireless network requires coded stream capable of adapting to dynamic network conditions instantly. To meet this requirement, various scalable video coding schemes have been developed, among which the Scalable Video Coding (SVC) extension of the H.264/AVC is the most recent one. In comparison with the scalable profiles of previous video coding standards, the SVC achieves significant improvement on coding efficiency performance. For adapting to dynamic network bandwidth, the SVC employs inter-layer switching between different temporal, spatial or/and fidelity layers, which is currently supported with instantaneous decoding refresh (IDR) access unit. However, for real-time adaptability, the SVC has to frequently employ the IDR picture, which dramatically decreases the coding efficiency. Therefore, an extension of SP picture from the AVC to the SVC for an efficient inter-layer switching is investigated and presented in this paper. Simulations regarding the adaptability to dynamic network bandwidth are implemented. Results of experiment show that the SP picture added SVC provides an average 1.2 dB PSNR enhancement over the current SVC while providing similar adaptive functionality.