• 방송공학회논문지
• /
• 제7권1호
• /
• pp.55-65
• /
• 2002

웨이브렛 변환은 최근에 동영상 처리 분야에서 비정체적 동영상 신호를 나타낼 수 있기 때문에 많은 관심을 받고 있다. 대역 분할 방식의 웨이브렛 변환을 사용한 움직임 예측은 많은 곳에서 응용되어 지고 있으나 움직임 예측 오류가 최저 주파수 대역에서 발생하면 움직임 예측 오류는 다음 단계로 누적되고 각각의 단계에서 계산에 소요되는 시간과 데이터량이 증가하는 문제점을 가지고 있다. 다른 한편으로 왜곡율 관점에서 주어진 비트율에서 최고의 화질을 얻는 웨이브렛 패킷이 제안되었다. 그러나 이 방법에서 웨이브렛 패킷 설계를 위해 많은 시간이 소요되는 단점이 존재한다. 그래서 소요시간 문제를 해결하기 위해 Top-down 방식이 제시되었으나 주어진 비트율에서 최적의 해를 찾지는 못했다. 본 논문에서는 영상의 분산이 영상의 복잡도를 나타낼 수 있다는 것을 고려했다. 그래서 동영상 압축을 위한 적응적인 웨이브렛 패킷을 사용한 고속의 다해상도 움직임 예측 방법을 제안하였다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 학술대회 논문집 정보 및 제어부문
• /
• pp.165-168
• /
• 2004

This paper propose a target tracker using phase correlation. The tracker consist of a pre-processing module, a translation estimation module based on phase correlation, a fine motion estimation module applied when confidence rate could not fulfill a threshold value and a reference image update module. The fine motion estimation module measure the shift, rotation and scale of input image compared to reference using Fourier-Mellin transform. Proposed tracker was tested its accuracy and robustness using some real indoor and outdoor image sequences.

• 응용통계연구
• /
• 제23권4호
• /
• pp.621-633
• /
• 2010

This paper considers a certain expense structure where a vendor of equity-linked investment product will collect its expenses continuously from the investor's account whenever the investment performance exceeds a certain threshold level. Under the Black-Scholes framework, we derive compact convolution formulas for evaluating the total expenses to be collected during the investment period by using the joint Laplace transform of the Brownian motion and its excursion time. We provide numerical examples for illustration.

• ETRI Journal
• /
• 제43권3호
• /
• pp.538-548
• /
• 2021

This paper proposes an approach to improve the performance of no-reference video quality assessment for sports videos with dynamic motion scenes using an efficient spatiotemporal model. In the proposed method, we divide the video sequences into video blocks and apply a 3D shearlet transform that can efficiently extract primary spatiotemporal features to capture dynamic natural motion scene statistics from the incoming video blocks. The concatenation of a deep residual bidirectional gated recurrent neural network and logistic regression is used to learn the spatiotemporal correlation more robustly and predict the perceptual quality score. In addition, conditional video block-wise constraints are incorporated into the objective function to improve quality estimation performance for the entire video. The experimental results show that the proposed method extracts spatiotemporal motion information more effectively and predicts the video quality with higher accuracy than the conventional no-reference video quality assessment methods.

• 한국수학교육학회지시리즈B:순수및응용수학
• /
• 제16권4호
• /
• pp.369-382
• /
• 2009

In this paper we establish various relationships among the generalized integral transform, the generalized convolution product and the first variation for functionals in a Banach algebra S($L_{a,b}^2$[0, T]) introduced by Chang and Skoug in [14]. We then derive an inverse integral transform and obtain several relationships involving inverse integral transforms.

• 대한수학회지
• /
• 제49권5호
• /
• pp.1065-1082
• /
• 2012

In this paper we first investigate the existence of the generalized Fourier-Feynman transform of the functional F given by $$F(x)={\hat{\nu}}((e_1,x)^{\sim},{\ldots},(e_n,x)^{\sim})$$, where $(e,x)^{\sim}$ denotes the Paley-Wiener-Zygmund stochastic integral with $x$ in a very general function space $C_{a,b}[0,T]$ and $\hat{\nu}$ is the Fourier transform of complex measure ${\nu}$ on $B({\mathbb{R}}^n)$ with finite total variation. We then define two sequential transforms. Finally, we establish that the one is to identify the generalized Fourier-Feynman transform and the another transform acts like an inverse generalized Fourier-Feynman transform.

• 대한전자공학회논문지SP
• /
• 제39권2호
• /
• pp.87-95
• /
• 2002

반화소 단위 움직임 추정(half pixel accuracy motion estimation, HPAME)과 블록 분류(block classification)를 이용한 계층적 고속 움직임 추정 알고리듬을 제안하였다 제안한 알고리듬은 기존의 MRME(multi-resolution motion estimation)알고리듬보다 우수한 화질을 유지하면서 계산량 및 비트량을 크게 줄일 수 있는 장점을 갖는다. 제안한 알고리듬에서는 다해상도 영상에 대한 움직임 추정 시 고주파 부대역의 움직임 추정에 기준 움직임으로 사용되는 기저대역의 움직임 벡터를 정확하게 추정하기 위하여 HPAME을 행한다. 그리고 고주파 부대역에서는 기저대역에서의 HPAME로 인한 계산량 및 비트량의 증가를 보상하기 위하여 움직임 추정이 필요한 블록들에 대하여서만 선별적으로 미소 움직임을 추정한다. 이때 고주파 부대역에서의 미소 움직임 추정의 수행 여부는 대응되는 기저대역 블록의 움직임 벡터 특성과 블록 분류에 따른 클래스 정보를 이용하여 결정한다 제안한 알고리듬의 성능은 컴퓨터 모의 실험 결과로부터 확인하였다.

• Journal of Multimedia Information System
• /
• 제4권4호
• /
• pp.225-232
• /
• 2017

Understanding actions in videos is an important task. It helps in finding the anomalies present in videos such as fights. Detection of fights becomes more crucial when it comes to sports. This paper focuses on finding fight scenes in Hockey sport videos using blur & radon transform and convolutional neural networks (CNNs). First, the local motion within the video frames has been extracted using blur information. Next, fast fourier and radon transform have been applied on the local motion. The video frames with fight scene have been identified using transfer learning with the help of pre-trained deep learning model VGG-Net. Finally, a comparison of the methodology has been performed using feed forward neural networks. Accuracies of 56.00% and 75.00% have been achieved using feed forward neural network and VGG16-Net, respectively.

• 대한전자공학회논문지
• /
• 제25권12호
• /
• pp.1513-1521
• /
• 1988

BMA, which is generally adopted in low bit-rate motion-compensated coding, performs properly under an assumption of rigid-body motion of moving objects. Since, however, the assumption can not be held in practical coding , the prediction errors with low correlation are generated. For effective transform codings of the interframe prediction errors, we propose a new transform coding technique which decomposes the prediction errors into two sources and then transforms them with DCT and WHT consecutively. The performance of the proposed algorithm is compared to those of the two conventional algorithms in terms of bit rate and subjective image quality.

• Structural Engineering and Mechanics
• /
• 제27권2호
• /
• pp.173-197
• /
• 2007

A wavelet-based procedure to generate artificial accelerograms compatible with a prescribed seismic design spectrum is described. A procedure to perform a baseline correction of the compatible accelerograms is also described. To examine how the frequency content of the modified records evolves with time, they are analyzed in the time and frequency using the wavelet transform. The changes in the strong motion duration and input energy spectrum are also investigated. An alternative way to match the design spectrum, termed the "two-band matching procedure", is proposed with the objective of preserving the non-stationary characteristics of the original record in the modified accelerogram.