• 응용통계연구
• /
• 제23권1호
• /
• pp.151-165
• /
• 2010

영상은 잡음센서이나 채널 전송에러에 의해 생기는 임펄스 잡음에 의해 자주 오염된다. 본 논문은 영상에서 이런 임펄스 잡음을 제거하는 방법에 대해 논의하고자 한다. 제안된 잡음제거는 SVM(Support Vector Machine)과 개선된 Adaptive Median 필터에 의해 이루어진다. SVM에 의해 영상에서 잡음픽셀여부를 검출하고 검출된 잡음픽셀은 개선된 Adaptive Median 필터에 의해 새로운 픽셀값으로 대체한다. 제안된 방법의 성능을 평가하기 위해 영상 실험을 통하여 salt-and-pepper 임펄스 잡음과 random-valued 임펄스 잡음을 고려하여 기존의 잡음제거 방법들과 정성적이고 MAE, PSNR를 통한 정량적인 비교를 하였다. 실험결과 제안된 방법은 잡음 제거와 미세한 부분에 대한 보존력이 뛰어나고 특히, 많이 오염된 영상에 대해서도 상당한 잡음제거 성능을 보였다.

• 한국음향학회지
• /
• 제8권4호
• /
• pp.53-59
• /
• 1989

본 논문에서는 지금까지 연구 되어온 고역통과 필터및 적응 transversal LMS필터와는 달리 적응 격자 위너 필터를 이용하여 적응 잡음 제거기를 구성함으로써 폐음과 심음을 분리할 수 있는 새로운 방법을 제시하였다. 이를 위하여 실제로 폐음및 심전도 신호를 검출하였으며, 제2의 심음 제거 방법으로 T파검출 알고리즘을 이용하여 T파위치를 측정하였다. 실험결과, 분리를 위하여 적응 transversal LMS는 100-200차, 적응 transversal MLMS(modified LMS) 필터는 75-100차, 적응 격자 위너 필터는 10-20차가 필요하였고, 적응 펄터링이 고역통과 필터링보다 저주파 신호성분의 손실이 없는 것으로 정확도가 향상되었다.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제51권10호
• /
• pp.468-476
• /
• 2002

Digital images are easily corrupted by noise during the data transmission, data capture and data processing. A technical method of noise analyzing and adaptive filtering for reducing of quantum noise in medical radio-fluoroscopy images is presented. By adjusting the characteristics of the filter according to local statistics around each pixel of the image as moving windowing, it is possible to suppress noise sufficiently while preserve edge and other significant information required in diagnosis. We proposed adaptive weighed median(AWM) filters based on local statistics. We showed two ways of realizing the AWM filters. One is a simple type of AWM filter, which is constructed by Homogeneous factor(HF). Homogeneous factor(HF) from the noise models that enables the filter to recognize the local structures of the image is introduced, and an algorithm for determining the HF fitted to the diagnostic systems with various inner statistical properties is proposed. We show by the experimented that the performances of proposed method is superior to these of other filters and models in preserving small details and suppressing the noise at homogeneous region. The proposed algorithms were implemented by Visual C++ language on a IBM-PC Pentium 550 for testing purposes and the effects and results of the filter in the various levels of noise and images were proposed by comparing the values of NMSE(normalized mean square error) with the value of the other existing filtering methods.

• 한국컴퓨터정보학회논문지
• /
• 제29권2호
• /
• pp.101-108
• /
• 2024

AI-OCR은 광학 문자 인식(OCR) 기술과 Artificial intelligence(AI)의 결합으로 사람의 인식이 필요하던 OCR의 단점을 보완하는 기술 향상을 이뤄내고 있다. AI-OCR의 성능을 높이기 위해서는 다양한 학습데이터의 훈련이 필요하다. 하지만 이미지 색상이 비슷한 밝기를 가진 경우에는 인식률이 떨어지기 때문에, Homomorphic filtering(HF)을 이용한 전처리 과정으로 색상 차이를 분명하게 하여 텍스트 인식률을 높이게 된다. HF은 감마값을 이용해 이미지의 고주파와 저주파를 각각 조절한다는 점에서 텍스트 추출에 적합하지만 감마값의 조절이 수동적으로 이뤄지는 단점이 존재한다. 본 연구는 시험적 과정을 거쳐 이미지의 대비, 밝기 및 엔트로피를 근거하는 감마의 임계값 범위를 제안한다. 제안된 감마값 범위를 적용한 HF의 실험 결과는 효율적인 AI-OCR의 높은 등장 가능성을 시사한다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 추계종합학술대회 논문집
• /
• pp.1159-1162
• /
• 1999

The ECG signal is very important information for diagnosis of patient and a cardiac disorder. It is hard to remove the noise because that is mixed with a lot of noise, and the error of the filtering will distort the ECG signal. The existing method for the filtering of the ECG signal has structure that has many steps for filtering, so that structure is complex and the processing speed is slow. For the improvement of that problem, we propose the method of filtering that has simple structure using the RBF neural networks and have good results.

• Journal of Communications and Networks
• /
• 제9권2호
• /
• pp.185-191
• /
• 2007

Alpha stable distribution is better for modeling impulsive noises than Gaussian distribution in signal processing. This class of process has no closed form of probability density function and finite second order moments. In general, Wiener filter theory is not meaningful in S$\alpha$SG environments because the expectations may be unbounded. We proposed a new adaptive recursive least p-norm Kalman filtering algorithm based on least p-norm of innovation process with infinite variances, and a new robust multiuser detection method based on least p-norm Kalman filtering. The simulation experiments show that the proposed new algorithm is more robust than the conventional Kalman filtering multiuser detection algorithm.

• 한국정보통신설비학회:학술대회논문집
• /
• 한국정보통신설비학회 2008년도 정보통신설비 학술대회
• /
• pp.141-145
• /
• 2008

Wireless sensor network (WSN) is expected to be used in many applications. However, sensor nodes still have some secure problems to use them in the real applications. They are typically deployed on open, wide, and unattended environments. An adversary using these features can easily compromise the deployed sensor nodes and use compromised sensor nodes to inject fabricated data to the sensor network (false data injection attack). The injected fabricated data drains much energy of them and causes a false alarm. To detect and drop the injected fabricated data, a filtering-based security method and adaptive methods are proposed. The number of different partitions is important to make event report since they can make a correctness event report if the representative node does not receive message authentication codes made by the different partition keys. The proposed methods cannot guarantee the detection power since they do not consider the filtering scheme. We proposed clustering method for filtering-based secure methods. Our proposed method uses fuzzy system to enhance the detection power of a cluster.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.941-946
• /
• 2005

In this paper an active random noise control using adaptive learning rate neural networks is presented. The adaptive learning rate strategy increases the learning rate by a small constant if the current partial derivative of the objective function with respect to the weight and the exponential average of the previous derivatives have the same sign, otherwise the learning rate is decreased by a proportion of its value. The use of an adaptive learning rate attempts to keep the learning step size as large as possible without leading to oscillation. It is expected that a cost function minimize rapidly and training time is decreased. Numerical simulations and experiments of active random noise control with the transfer function of the error path will be performed, to validate the convergence properties of the adaptive learning rate Neural Networks. Control results show that adaptive learning rate Neural Networks control structure can outperform linear controllers and conventional neural network controller for the active random noise control.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2014년도 하계학술대회
• /
• pp.179-180
• /
• 2014

The main goal of this paper is to present an adaptive filter system using NLMS(Normalized Least mean square) adaptive algorithm for noise cancellation. The proposed algorithm has less computational complexity and better convergence property than the former algorithms like spectral subtraction algorithm, etc. We use TIMIT criterion voice and Noisex-92 for the experiment. The experimental result shows the feasibility of our algorithm for filtering noise from voice effectively.

• 대한전자공학회논문지
• /
• 제24권6호
• /
• pp.1005-1012
• /
• 1987

An adaptive scalar DCT-Wiener filtering method that can be applied to the problem of restoring an image degraded by white Gaussian noise is presented. In this method, the power spectrum needed to Winer filering is adaptively estimated in block-wise according to the lical properties in transform domain. In addition, overlapping method for reducing the block artifact is considered. Experimental results show that the adaptive Wiener filter by the proposed method yields performance improvement and better image quality over the nonadaptive one and the spatial Lee filter.