• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2016년도 추계학술대회
• /
• pp.965-968
• /
• 2016

본 논문은 PDF 웨이블릿 변환과 TVG 필터를 이용한 Sonar Image 의 스펙클 노이즈 제거에 관한 연구이다. TVG 필터는 해저면 바닥의 불규칙한 물체를 식별하는데 방해되는 Speckle Noise를 시간적이고 귀납적인 관찰의 결과로 얻어진 Gain 값으로 제거 한다. 실험 결과, 제안된 필터는 Speckle Noise를 약 90% 제거하여 보다 향상된 해저 영상 식별이 가능하다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 제13회 신호처리 합동 학술대회 논문집
• /
• pp.679-682
• /
• 2000

In this paper, a MMSE filter in homomorphic wavelet transform domain is proposed for restoring an ultrasonic images corrupted by speckle noise. In order to remove effectively the speckle noise which is a kind of multiplicative noise, speckle noise is transformed into a form of additive noise and then the additive noise is denoised through the MMSE filter in homomorphic wavelet transform domain. The proposed method shows much higher quality in terms of ISNR and subject quality.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2009년도 춘계학술대회
• /
• pp.156-159
• /
• 2009

Speckle noise reduction for 3D power doppler ventricle coherent image for restoration and enhancement using wavelet packet transform with separated thresholding is presented. Wavelet Packet Transform divide into low frequency component image to high frequency component image to be multi-resolved. speckle noise is located on high frequency component in multiresolution image mainly. A ventricle image is transformed and inversed with separated threshold function from low to high resolved images for restoration to be utilize visualization for ventricle diagnosis. The experimental result shows that the proposed method has better performance in comparison with the conventional method.

• ETRI Journal
• /
• 제41권1호
• /
• pp.32-41
• /
• 2019

We propose a speckle reduction technique in electronic holographic display systems, where digital micro-mirror array devices are used as spatial light modulators. By adopting a programmable filtration in a general 4-f optic configuration, it is shown that the signal spectrum components in the frequency domain of a viewing-window-based holographic display system can be selectively filtered. Compared to the widely utilized single-sideband filtration techniques in electronic holographic display systems, our proposed programmable filtration can be utilized to effectively reduce the speckles in the reconstruction of point-cloud-based computer-generated holograms. Experimental results are presented to verify our proposed concept.

• 대한의용생체공학회:의공학회지
• /
• 제37권4호
• /
• pp.140-146
• /
• 2016

The ultrasound imaging in medical diagnosis has become a popular modality because of its safe, noninvasive, portable, relatively inexpensive, and provides a real-time image formation. However, usefulness of ultrasound imaging is at times limited due to the presence of signal-dependent noise like as speckle. Therefore, noise reduction is very important, as various types of noise generated limits the effectiveness of medical image diagnosis. This paper introduces a speckle noise reduce algorithm using total variation denoising (TVD) in Laplacian pyramid. With this method, speckle is removed by TVD of bandpass ultrasound images in Laplacian pyramid domain. For TVD in each pyramid layer, a ${\lambda}$ is selected by trial-and-error method. The visual comparison of despeckled 'in vivo' ultrasound images from pancreas shows that the proposed method could effectively preserve edges and detailed structures while thoroughly suppressing speckle. For a Simulated B-mode image, contrast-to-noise-ratio (CNR) and signal-to-noise-ratio (SNR) were obtained like 4.65 dB and 14.11 dB, respectively. The results show that the proposed method can conduct better than some of the existing methods in terms of the CNR and the SNR.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1990년도 추계학술대회
• /
• pp.5-10
• /
• 1990

The object detection capabilities of ultrasonic imaging systems are limited by the ability of the detection process to distinguish the resolved object signals from backscattered speckle noise. It has been shown that the phase component of the Fourier transform of the speckle noise is random. Based on this property. we propose a new algorithm for distinguishing between speckle and specular targets. The proposed algorithm is implemented by taking the Fourier transform of the received signal, low-pass filtering the phase, and taking the inverse Fourier transform of the filtered phase to enhance specular reflectors and reduce speckle in the image. Simulations and experiments using phantoms confirm the algorithm yielding significant reduction of speckle noise.

• 한국인터넷방송통신학회논문지
• /
• 제9권6호
• /
• pp.155-161
• /
• 2009

SAR영상의 가장 큰 문제점은 경계선 부근에서 스패클(Speckle)잡음을 어떻게 줄이느냐 하는 것이다. 본 논문에서는 제안한 방법을 이용하여 경계선을 보존할 수 있는 효과적인 필터를 개발하고자 한다. 스패클 잡음을 줄이면서 에지 영역에 대한 블러링 없는 영상을 추출하기 위하여 웨이브렛 기반의 sigma 필터를 적용하였다. 실험 결과 에지정보에 대한 블러링을 줄인 출력 영상을 구성하였다. 제안한 방법을 미디언 필터와 비교한 결과, 스패클 잡음을 효과적으로 제거한 우수한 영상을 얻을 수 있었다.

• 한국광학회지
• /
• 제5권2호
• /
• pp.217-224
• /
• 1994

반점 간섭무늬 위상단면도의 잡음을 제거하기 위해 위상이동 convolution과 2-bit 양자화 스무딩 필터를 조합하였다. 위상이동 convoltion처리로 톱니파 형태를 갖는 반점 간섭무늬 위상단면도에서 2.pi. jump 양단의 edge 성분을 완벽하게 유지하면서 스무딩 처리를 할 수 있었다. 2-bit 양자화 스무딩 필터는 반점 간섭무늬 위상단면도의 스무딩 처리를 하는데 있어서 average, low-pass 필터 및 median 필터 보다 처리속도가 빠르고 S/N비 특성이 우수하였다. 잡음을 제거한 반점 위상단면도를 path dependent unwrapping 알고리즘을 적용하여 위상의 2.pi. 불연속성을 해소하고 펼쳐 보았다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2008년도 International Symposium on Remote Sensing
• /
• pp.185-187
• /
• 2008

Speckle noise is primarily due to the phase fluctuations of the electromagnetic return signals. Since inherent spatial-correlation characteristics of speckle in SAR images are not exploited in existing multiplicative models for speckle noise. In this paper, we investigate the efficiency of the filters: Kuan, Gamma, Enhanced Lee, and Enhanced Frost filters in reduction of speckle noise in forestry area. Selection of the suitable filter can be played an important role in applications such as estimation of biomass in forestry areas. We employ ALOSPALSAR image use L-band frequency for this investigation. The filters test on different sites in PALSAR image of the northern forests part of Iran.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 1999년도 Proceedings of International Symposium on Remote Sensing
• /
• pp.144-147
• /
• 1999

The objective of this study is to quantitatively evaluate the effects of various SAR speckle reduction methods for multisource land-cover classification by backpropagation neural network, especially over the coastal region. The land-cover classification using neural network has an advantage over conventional statistical approaches in that it is distribution-free and no prior knowledge of the statistical distributions of the classes is needed. The goal of multisource land-cover classification acquired by different sensors is to reduce the classification error, and consequently SAR can be utilized an complementary tool to optical sensors. SAR speckle is, however, an serious limiting factor when it is exploited for land-cover classification. In order to reduce this problem. we test various speckle methods including Frost, Median, Kuan and EPOS. Interpreting the weights about training pixel samples, the “Importance Value” of each SAR images that reduced speckle can be estimated based on its contribution to the classification. In this study, the “Importance Value” is used as a criterion of the effectiveness.