• 전자공학회논문지
• /
• 제54권5호
• /
• pp.42-47
• /
• 2017

본 논문에서는 디지털 영상의 배포에서, 위 변조에 사용되는 미디언 필터링 (Median Filtering : MF)을 분류하는 포렌식 검출 알고리즘을 제안한다. 이러한 문제를 해결하기 위한 특징벡터는 영상의 에지 검출량 정보 32, 64, 128에 대한 허프변환(Hough Transform)에 의하여, 각 허프라인 (Hough Line)의 양끝 좌표값과 Angle-Distance 좌표상의 허프픽크치 (Hough Peaks)를 조합하여 42-Dim.으로 구성하였다. 변조된 영상들 중에서 미디언 필터링을 분류하는 검출기는 SVM (Support Vector Machine)에서 특징벡터를 학습하여 구현되었다. 제안된 미디언 필터링 검출 알고리즘은 특징벡터의 길이가 10-Dim.의 MFR (Median Filtering Residual) 스킴 및 686-Dim.의 SPAM (Subtractive Pixel Adjacency Matrix) 스킴과 비교하여 원영상, 평균필터링 ($3{\times}3$), JPEG (QF=90, 70) 압축, 가우시안 필터링 ($3{\times}3$, $5{\times}5$) 영상 모두에서 미디언 필터링의 포렌식 분류율이 99% 이상의 성능을 확인하였다.

• 대한전자공학회논문지SP
• /
• 제38권4호
• /
• pp.411-421
• /
• 2001

일반화 대칭 변환 (generalized symmetry transform, GST)은 주어진 영상에서 사전 분할이 없이 국부성과 반사 대칭성을 결합하여 대칭을 측정하고 관심 영역을 추출한다. GST의 거리 가중치 함수에서 국부적인 대칭성이 반영되며 이 함수의 표준 편차 u에 의해 GST의 수행 범위가 조절된다. 넓은 관심영역을 추출하기 위해 반지름 r이 큰 검색영역 내에서의 대칭성이 추출될 필요가 있다. 이에 따라서 GST의 수행시간은 r에 따라 2차적으로 증가하게 된 본 논문에서는 이를 개선하기 위해 선택적 방향 주의를 가지는 수정된 스캔라인 GST를 제안한다. 제안된 GST는 기존의 GST와 유사한 대칭 특성을 추출하지만 선택적 방향의 기울기만을 고려한 스캔라인 위의 에지 화소쌍에서 GST를 수행함으로써 r에 따라서 이의 수행시간이 선형적으로 증가된다 특히 r이 큰 경우에 선택적 방향에 대해서만 적용하면 기존의 GST의 계산량이 비대해지는 단점을 보완해 줄 수 있다. 제안된 GST가 기존의 GST보다 시간적으로 효과적이며 유용하다는 것이 여러 종류의 영상에 대한 실험으로 확인되었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권10호
• /
• pp.4797-4813
• /
• 2018

The transform domain communication system (Hereinafter referred to as TDCS) takes on numerous advantages, inclusive of anti-jamming and low probability of detection. Yet its application is confined by the consistent spectrum in the transmitter and receiver, which is not possible in the case of a huge distance exsits between them. In this paper, a TDCS based modified weighted fractional fourier transform (WFRFT) is proposed to solve the problem resulting from spectrum mismatching for TDCS application. The amplitude and phase information are incorporated with the TDCS signals and transmit to the receiver together in the wake of a modified WFRFT. The basic function and the TDCS signals shall be accessible to the receivers in the wake of an inverse WFRFT transform, which make sure that the original information can be demodulated properly. The system's reliability while transmitting signals with different modulation methods and with spectrum mismatching is demonstrated by bit error rate (BER). In the meantime, the constellations of the signals and the BER performances at the eavesdropper demonstrate the proposed system is better secured.

• Smart Structures and Systems
• /
• 제18권5호
• /
• pp.1063-1085
• /
• 2016

Taking advantage of the high sensitivity and long-distance detection capability of acoustic emission (AE) technique, this paper focuses on the crack detection in rail head, which is one of the most vulnerable parts of rail track. The AE source location and noise cancellation were studied on the basis of practical rail profile, material and operational noise. In order to simulate the actual AE events of rail head cracks, field tests were carried out to acquire the AE waves induced by pencil lead break (PLB) and operational noise of the railway system. Wavelet transform (WT) was first utilized to investigate the time-frequency characteristics and dispersion phenomena of AE waves. Here, the optimal mother wavelet was selected by minimizing the Shannon entropy of wavelet coefficients. Regarding the obvious dispersion of AE waves propagating along the rail head and the high operational noise, the wavelet transform-based modal analysis location (WTMAL) method was then proposed to locate the AE sources (i.e. simulated cracks) respectively for the PLB-induced AE signals with and without operational noise. For those AE signals inundated with operational noise, the Hilbert transform (HT)-based noise cancellation method was employed to improve the signal-to-noise ratio (SNR). Finally, the experimental results demonstrated that the proposed crack detection strategy could locate PLB-simulated AE sources effectively in the rail head even at high operational noise level, highlighting its potential for field application.

• 대한의용생체공학회:의공학회지
• /
• 제20권3호
• /
• pp.265-274
• /
• 1999

본 논문에서는 3차원 의료 영상에 적합한 interframe 부호화 알고리즘을 제안하였다. 제안된 알고리즘은 움직임 추정/보상과 변환 부호화를 사용하는 동영상 압축 방법에 기초를 두고 있다. 이 알고리즘에서는 움직임 추정을 위해 와핑 방법이 채택되었다. 그리고, 부호화 효율을 향상시키기 위해 적응적 모드 선택을 사용하여 웨이블릿 변환 영역에서 움직임 조상된 오차 영상과 원 영상을 혼합한다. 이렇게 혼합된 영상은 제로트리 부호화 기법에 의해 압축된다. 본 논문은 제안된 방법이 3차원 의료 영상의 부호화에 매우 유용함을 증명하였다. 모의실험결과는 제안된 방법이 슬라이스 간의 간격과 무관하게 항상 좋은 결과를 얻을 수 있도록 하고, 그럼으로써 3차원 의료 영상 압축에 매우 바람직하다는 사실을 보여준다.

• 한국정보과학회논문지:시스템및이론
• /
• 제37권6호
• /
• pp.323-329
• /
• 2010

문자열에 대한 편집 거리 문제는 하나의 문자열을 다른 문자열로 변환할 때 필요한 최소한의 연산의 개수를 구하는 문제이다. 영어와 같은 1차원 문자열에 대한 최적해에 대해서는 오랫동안 연구가 진행되어 왔으나, 한글과 같이 좀 더 복잡한 언어에 대한 편집 거리에 대해서는 많은 연구가 진행되지 못했다. 본 논문에서는 음소와 음절을 구분하여 편집거리를 구하는 기존 연구를 확장하여, 음소간의 유사도를 정의하고 이를 이용하여 유사한 단어를 더 정확하게 구분해 내는 알고리즘을 제안한다.

• 한국멀티미디어학회논문지
• /
• 제17권2호
• /
• pp.113-123
• /
• 2014

Electrophysiological recordings are considered a reliable method of assessing a person's alertness. Sleep medicine is asked to offer objective methods to measure daytime alertness, tiredness and sleepiness. As EEG signals are non-stationary, the conventional method of frequency analysis is not highly successful in recognition of alertness level. In this paper, EEG signals have been analyzed using wavelet transform as well as discrete wavelet transform and classification using statistical classifiers such as euclidean and mahalanobis distance classifiers and a promising method SVM (Support Vector Machine). As a result of simulation, the average values of accuracies for the Linear Discriminant Analysis (LDA)-Quadratic, k-Nearest Neighbors (k-NN)-Euclidean, and Linear SVM were 48%, 34.2%, and 86%, respectively. The experimental results show that SVM classification method offer the better performance for reliable classification of the EEG signal in comparison with the other classification methods.

• 전기학회논문지
• /
• 제57권8호
• /
• pp.1311-1317
• /
• 2008

In this paper, a fault location algorithm using wavelet transform is proposed for HVDC cable lines. The arriving instants of the first and second fault-induced backward travelling waves can be detected by using wavelet transform. The fault distance is estimated by using the time difference between the two instants of backward travelling waves and the velocity of the travelling wave. To distinguish between the backward wave from fault point and the backward wave from the remote end, polarities of backward waves are used. The proposed algorithm is verified varying with fault distances and fault resistances in underground cables of VSC(voltage source converter) HVDC system and CSC(Current Source Converter) HVDC respectively. Performance evaluations of the proposed algorithm shows that it has good ability for a fault location of HVDC cable faults.

• 한국해양공학회지
• /
• 제22권1호
• /
• pp.30-36
• /
• 2008

Underwater ambient noise originating from geophysical, biological, and man-made acoustic sources contains information on the source and the ocean environment. Such noise affectsthe performance of sonar equipment. In this paper, three steps are used to identify the ambient noise source, detection, feature extraction, and similarity measurement. First, we use the zero-crossing rate to detect the ambient noisesource from background noise. Then, a set of feature vectors is proposed forthe ambient noise source using the Hilbert-Huang transform and the Karhunen-Loeve transform. Finally, the Euclidean distance is used to measure the similarity between the standard feature vector and the feature vector of the unknown ambient noise source. The developed algorithm is applied to the observed ocean data, and the results are presented and discussed.

• Structural Engineering and Mechanics
• /
• 제83권1호
• /
• pp.67-77
• /
• 2022

In this study, frictionless continuous and discontinuous contact problems of a magneto-electro-elastic layer in the presence of the body force were discussed. The layer was indented by a rigid cylindrical insulating punch and supported by a rigid substrate without bond. Applying the Fourier integral transform technique, the general expressions of the problem were derived in the presence of body force. Thanks to the boundary conditions, the singular integral equations were obtained for both the continuous and the discontinuous contact cases. Gauss-Chebyshev integration formulas were used to transform the singular integral equations into a set of nonlinear equations. Contact width under the punch, initial separation distance, critical load, separation regions and contact stress under the punch and between the layer, and substrate were given as a result.