• 한국통신학회논문지
• /
• 제29권10C호
• /
• pp.1414-1424
• /
• 2004

본 논문에서는 동영상의 장면전환점 중 급진적인 장면전환점인 컷(cut)과 점진적인 장면전환점인 페이드(fade)와 디졸브(dissolve) 구간을 웨이블릿 변환영역에서 검출하는 알고리즘을 제안한다. 웨이블렷 변환을 이용한 기존의 연구들은 공간영역과 변환영역 각각의 특징을 이용하여 장면전환점을 검출한다. 그러나 본 논문은 입력된 컬러영상을 먼저 YW 공간으로 변환하고, Y 성분에 대해 리프팅기법을 적용하여 2 레벨 웨이블릿 변환 후, 변환영역에서 공간영역의 특징이 유지되는 저주파 부대역을 히스토그램 비교하고, 나머지 고주파 부대역에서 추출된 에지 정보를 전체(global), 부분(semi-global), 국부(local) 영역으로 정의하여 웨이블릿 에지 히스토그램 비교를 한다. 모의실험 결과 기존의 방법보다 recall에서는 약 17%, precision에서는 약 18%의 성능향상을 보였으며 점진적인 장면 전환점인 페이드와 디졸브 구간 검출에도 좋은 성능을 나타내었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제6권9호
• /
• pp.2424-2441
• /
• 2012

In this paper, we propose a fake iris detection method that combines the optical and textural features of the human eye. To extract the optical features, we used dual Purkinje images that were generated on the anterior cornea and the posterior lens surfaces based on an analytic model of the human eye's optical structure. To extract the textural features, we measured the amount of change in a given iris pattern (based on wavelet decomposition) with regard to the direction of illumination. This method performs the following two procedures over previous researches. First, in order to obtain the optical and textural features simultaneously, we used five illuminators. Second, in order to improve fake iris detection performance, we used a SVM (Support Vector Machine) to combine the optical and textural features. Through combining the features, problems of single feature based previous works could be solved. Experimental results showed that the EER (Equal Error Rate) was 0.133%.

• 한국전기전자재료학회논문지
• /
• 제19권10호
• /
• pp.976-983
• /
• 2006

Multi-resolution Signal Decomposition (MSD) Technique of Wavelet Transform has interesting properties of capturing the embedded horizontal, vertical and diagonal variations within an image in a separable form. This feature was exploited to identify individual partial discharge sources present in multi-source PD pattern, usually encountered during practical PD measurement. Employing the Daubechies wavelet, feature were extracted from the third level decomposed and reconstructed horizontal and vertical component images. These features were found to contain the necessary discriminating information corresponding to the individual PD sources and multi-PD soruces.

• 한국정밀공학회지
• /
• 제22권4호
• /
• pp.37-43
• /
• 2005

Real time monitoring of exit burr formation is critical in manufacturing automation. In this paper, acoustic emission (AE) was used to detect the burr formation during drilling. By using wavelet transform (WT), AE data were compressed without unnecessary details. Then the transformed data were used as selected features (inputs) of a back-propagation artificial neural net (ANN). In order to validate the in process AE monitoring system, both WT-based ANN and cutting condition (cutting speed, feed, drill diameter, etc.) based ANN outputs were compared with experimental data.

• International Journal of Precision Engineering and Manufacturing
• /
• 제8권1호
• /
• pp.38-45
• /
• 2007

We propose a preprocessing method for removing floating particles from underwater images based on an analysis of the image features. We compared baseline JPEG and wavelet codec methods to determine the method best suited for underwater images. The proposed preprocessing method enhanced the compression ratio and resolution, and provided an efficient means of compressing the images. The wavelet codec method yielded better compression ratios and image resolutions. The results suggest that the wavelet codec method linked with the proposed preprocess method provides an efficient codec processor and transmission system for underwater images that are used for searches and transmitted via ultrasonic communications.

• 한국정보통신학회논문지
• /
• 제4권2호
• /
• pp.461-468
• /
• 2000

디지털 기술 발전에 따른 오디오 데이터의 증가는 여러 컴퓨터 응용에 사용되면서 데이터를 관리하고 사용하기 위해, 내용기반 질의와 유사성 검색과 같은 새로운 기능을 갖는 데이터베이스 시스템의 개발이 불가피하게 됐다. 내용 기반 질의를 위한 빠르고 정확한 검색은 이러한 응용 시스템들에 필요하다. 효율적인 내용기반 색인과 유사성 검색의 설계는 관련성 있는 데이터의 빠른 검색을 제공하기 위한 주된 요소이다. 본 논문에서는 소파(Wavelet) 변환을 이용한 한국 전통 음악 데이터베이스의 오디오 색인을 위한 방법을 제안한다. 또한 소파 변환을 이용해 오디오 데이터에 대한 색인의 가능성을 보인다.

• 한국복합재료학회:학술대회논문집
• /
• 한국복합재료학회 2002년도 추계학술발표대회 논문집
• /
• pp.151-154
• /
• 2002

The objective of this study is to find the change of signal characteristics of matrix cracks due to the different specimen shapes. As the concept of the smart structure, monitoring of acoustic emission (AE) can be applied to inspect the fracture of the structures in operating condition using built-in sensors. To understand the characteristics of matrix crack signals, we performed tensile tests by changing the thickness and width of the specimens. This paper describes the implementation of time-frequency analysis such as wavelet transform (WT) fur the quantitative evaluation of fracture signals. The experimental result shows the distinctive signal features in frequency domain due to the different specimen shapes.

• Journal of Advanced Marine Engineering and Technology
• /
• 제31권1호
• /
• pp.95-102
• /
• 2007

The issues of preventive and condition-based maintenance, online monitoring, system fault detection, diagnosis and prognosis are of increasing importance. This study introduces a technique to detect and identify faults in induction motors. Stator currents were measured and stored by time domain. The time domain is not suitable for representing current signals, so wavelet transform is used to convert the signals onto frequency domain. The raw signals can not show the significant feature, therefore difference values between the signal of the health conditions and that of the fault conditions are applied. The difference values were transformed by wavelet transform and the features are extracted from the transformed signals. The dynamic time warping method was used to identify the fault type. This study describes the results of detecting fault using wavelet analysis.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2003년도 하계학술대회 논문집 C
• /
• pp.1879-1881
• /
• 2003

In this paper, acoustic signals in $SF_6$ gas were analyzed using wavelet transform. For this, the PD sources in the $SF_6$ gas were divided into corona discharge surface discharge void discharge and crossing particle and acoustic signals were used to detect the PD sources. The measured signals were time-frequency distribution by wavelet transform and the features were extracted from the PD sources. As a result the characteristics of the PD sources were different. And this results is going to be used for basis diagnosis of $SF_6$ gas insulated apparatus.

• Journal of the Optical Society of Korea
• /
• 제15권1호
• /
• pp.103-108
• /
• 2011

Recently reported asynchronous-optical-sampling terahertz (THz) time-domain spectroscopy enables high-resolution spectroscopy due to a long time-delay window. However, a long-lasting tail signal following the main pulse is often measured in a time-domain waveform, resulting in spectral fluctuation above a background noise level on a high-resolution THz amplitude spectrum. Here, we adopt the wavelet power spectrum estimation technique (WPSET) to effectively remove the spectral fluctuation without sacrificing spectral features. Effectiveness of the WPSET is verified by investigating a transmission spectrum of water vapor.