Wavelet Denoising Filter를 이용한 측위 정밀도 향상 기법 성능

신동수,박지호,박영식,황유민,김진영 통신위성ㆍ우주산업연구회 2014 한국위성정보통신학회논문지 Vol.9 No.3

최근, 현대전은 GPS 위치측위를 바탕으로 정밀타격체계 및 미사일 방어체계가 핵심이 되어가고 있다. 하지만 군 환경 특성상 산악지형 및 시가전에서의 지형지물로 인한 large/small scale fading, 주파수 간섭 등으로 인해 오차를 가진 위치정보를 얻게 된다. 이는아군 위치 파악 실패로 인한 지원 지연 및 유도탄 오폭으로 인명피해를 발생시키게 된다. 본 연구는 위치오차를 보정하기 위해wavelet denoising filter를 이용한 간섭완화 측위기법을 제안한다. 실험 결과는 본 연구실에서 수행한 GPS/QZSS/Wi-Fi밀결합 측위기법의 실증 테스트 결과와 wavelet denoising filter를 적용한 시스템의 시뮬레이션 결과로 간섭완화 성능을 나타낸다. Waveletdenoising filter를 적용한 시스템의 시뮬레이션 결과는 기존 GPS보다 평균 21.6% 의 정확도 향상을 보이며 제안한 시스템 모델의우수성을 입증한다. Recently, precision guided munition systems and missile defense systems based on GPS have been taking a key role inmodern warfare. In warfare however, unexpected interferences cause by large/small scale fading, radio frequencyinterferences, etc. These interferences result in a severe GPS positioning error, which could occur late supports and friendlyfires. To solve the problems, this paper proposes an interference mitigation positioning method by adopting a waveletdenoising filter algorithm. The algorithm is applied to a GPS/QZSS/Wi-Fi combined positioning system which was performedby this laboratory. Experimental results of this paper are based on a real field test data of a GPS/QZSS/Wi-Fi combinedpositioning system and a simulation data of a wavelet denoising filter algorithm. At the end, the simulation resultdemonstrates its superiority by showing a 21.6% improved result in comparison to a conventional GPS system.

Image Denoising Method based on Deep Learning using Improved U-net

Jaewook Han,Jinwon Choi,Changwoo Lee 대한전자공학회 2021 IEIE Transactions on Smart Processing & Computing Vol.10 No.4

Various methods, including block-matching and 3D filtering (BM3D), have been proposed for image denoising. Recently, studies on deep learning methods for image denoising have been on the rise. In this paper, we propose a new structure for a deep neural network that improves image denoising performance. Among the existing deep neural networks, we improve U-net, which is widely used for image restoration, through the inclusion of pre-processing and post-processing and by modifying each of its stages. Extensive simulations show that the proposed structure performs very well for a wide range of noise levels with a single trained parameter, and it exhibits superior image denoising performance compared to conventional deep neural networks.

Research on Image Nonlocal Denoising Algorithm based on Wavelet Decomposition

Jing Zhang 보안공학연구지원센터 2015 International Journal of Signal Processing, Image Vol.8 No.9

In the field of image denoising, nonlocal image denoising algorithm is a nonlinear, space average denoising algorithm, it will not cause boundary blurred, it is an effective denoising algorithm. But its application still has limitations because of it taking much longer time, in this paper, the method was improved, image signal can be divided into high frequency and low frequency part using wavelet decomposition, nonlocal denoising algorithm is used in low-frequency approximate signal, for high frequency detail signals using wavelet filtering method for denoising. The experimental results show that the method improves the speed of image processing and has good practical value.

GCST를 이용한 인간시각필터의 영상 잡음 제거

이적식,Lee, Juck-Sik 한국융합신호처리학회 2008 융합신호처리학회 논문지 (JISPS) Vol.9 No.4

영상향상 방법 중의 하나인 잡음제거는 공간영역과 변환영역 필터링에 대해서 많은 연구가 되어 왔다. 최근에는 에너지 집중도가 탁월하고 다분해능 성질을 갖는 웨이브릿 변환이 많이 사용되고 있다. 그러나 최종 사용자가 인간인 경우에는 인간시각체계에 기반한 변환을 사용하는 것이 시각적으로 유용하므로, 본 논문에서는 인간시각필터로 고려되는 Gabor 코사인과 사인 함수를 이용한 변환을 영상 잡음제거 분야에 적용하였다. 제안한 방법은 웨이브릿 변환과 다른 종류의 인간시각필터인 Gaussian 미분 변환에 대해서 피크신호대잡음비로 잡음제거 성능을 비교하였다. 여러 가지 잡음의 3가지 레벨에 대해서 실제 영상의 실험으로부터 제안한 변환이 BWT와 DGT보다 PSNR이 각각 0.41, 0.14dB 더 좋은 결과를 얻었다. Image denoising as one of image enhancement methods has been studied a lot in the spatial and transform domain filtering. Recently wavelet transform which has an excellent energy compaction and a property of multiresolution has widely used for image denoising. But a transform based on human visual system is visually useful if an end user is human beings. Therefore, Gabor cosine and sine transform which is considered as human visual filter is applied to image denoising areas in this paper. Denoising performance of the proposed transform is compared with those of the derivatives of Gaussian transform being another human visual filter and of discrete wavelet transform in terms of PSNR. With three levels of various noises, experimental results for real images show that the proposed transform has better PSNR performance of 0.41dB than DWT and 0.14dB than DGT.

스마트폰으로 촬영한 사진의 잡음제거를 위한 단일 이미지 이중 애버리징 기법

조훈민,이성길 한국정보과학회 2023 정보과학회논문지 Vol.50 No.1

Image-denoising algorithms have long been actively researched to remove noise generated in pixel signals. There is a denoising technique for a single image with Gaussian noise, a technique for removing noise using multiple photos taken by a fixed camera, and a technique for removing noise by learning the difference using deep learning. However, the noise in actual smartphone photographs does not have the same Gaussian distribution at each pixel, and taking multiple photos costs a lot of time. Deep learning disadvantages the ground truth image without noise is essential. Therefore, this paper analyzes the characteristic of noise appearing in images taken with smartphones and uses it for denoising. In addition, a single image containing noise is divided into several small areas, showing similar results to denoising using an average of multiple images. Accordingly, this technique can adequately perform denoising using a single noise image photographed by a smartphone without the ground truth image learning.

An Improved Spatially Selective Noise Filtration for Real-time Denoising of Acoustic Emission Signal

Jian Wang,Guangming Li,Peng Sun,Ruijuan Jiang,Yiyan Chen 보안공학연구지원센터 2015 International Journal of Signal Processing, Image Vol.8 No.7

The denoising of acoustic emission (AE) signal plays an important role in structural health monitoring. This paper proposes the improved spatially selective noise filtration (SSNF) which can eliminate the Gaussian white noise well. Firstly, through the comparison of different vanishing moments, “db5” is chose as the mother wavelet. And the Mallat algorithm is used in the composition and reconstruction of signal processing. Secondly, according to the signal noise ratio of wavelet reconstructed coefficients of AE signal, two coefficients are chosen to the next step. Lastly, the denoising algorithm uses the high degree of correlation between coefficients to realize the improved SSNF. Compared with the SSNF, the improved SSNF can avoid “glitches” and realize real-time denoising. And according to the simulation results, the improved SSNF can realize real-time denoising of AE signal.

Image Denoising with a Convolution Neural Network using Gaussian Filtered Residuals

Laavanya Mohan,Vijayaraghavan Veeramani 대한전자공학회 2021 IEIE Transactions on Smart Processing & Computing Vol.10 No.2

Deep learning using a convolutional neural network has become a state-of-art technique in image processing. In recent scenarios, image denoising using a residual image in deep learning has been popular. However, one aspect missing in these methods is that the residual image has all the noise and very small structured details of the input image. Therefore, we have developed a Gaussian filter residual convolutional neural network architecture for color image denoising. Gaussian residual learning was used to boost the denoising performance. The architecture is designed to remove additive white Gaussian noise, which is one of the most basic types of noise that affects an image when captured. The network with Gaussian residual learning removes the clean image using the features learned from the hidden layer. The peak signal-to-noise ratio and structural similarity index measure achieved by our method reveals that the presented approach is better at denoising images with Gaussian noise than a convolutional neural network.

Gamma spectrum denoising method based on improved wavelet threshold

Xie Bo,Xiong Zhangqiang,Wang Zhijian,Zhang Lijiao,Zhang Dazhou,Li Fusheng 한국원자력학회 2020 Nuclear Engineering and Technology Vol.52 No.8

Adverse effects in the measured gamma spectrum caused by radioactive statistical fluctuations, gamma ray scattering, and electronic noise can be reduced by energy spectrum denoising. Wavelet threshold denoising can be used to perform multi-scale and multi-resolution analysis on noisy signals with small root mean square errors and high signal-to-noise ratios. However, in traditional wavelet threshold denoising methods, there are signal oscillations in hard threshold denoising and constant deviations in soft threshold denoising. An improved wavelet threshold calculation method and threshold processing function are proposed in this paper. The improved threshold calculation method takes into account the influence of the number of wavelet decomposition layers and reduces the deviation caused by the inaccuracy of the threshold. The improved threshold processing function can be continuously guided, which solves the discontinuity of the traditional hard threshold function, avoids the constant deviation caused by the traditional soft threshold method. The examples show that the proposed method can accurately denoise and preserves the characteristic signals well in the gamma energy spectrum

Over blur를 감소시킨 Deep CNN 구현

이성훈,이광엽,정준모 한국전기전자학회 2018 전기전자학회논문지 Vol.22 No.4

In this paper, we have implemented a network that overcomes the over-blurring phenomenon that occurs whenremoving Gaussian noise. In the conventional filtering method, blurring of the original image is performed to removenoise, thereby eliminating high frequency components such as edges and corners. We propose a network that reducingover blurring while maintaining denoising performance by adding denoised high frequency components to denoisers basedon CNN. 본 논문에서, Gaussian noise를 제거할 때 발생하는 over blurring 현상을 감소시키는 network를 구현하였다. 기존 filtering방식은 원 영상을 blurring하여 noise를 제거함으로써, edge나 corner 같은 high frequency 성분도 함께 지워지는 것을 확인할 수 있다. CNN (Convolutional Neural Network)기반 denoiser의 경우도 사소한 edge, keypoint를 noise로 인식하여 이러한정보를 잃게 된다. 우리는 CNN을 기반으로 denoising된 high frequency 성분만을 획득하여 기존 denoiser에 추가함으로써denoising 성능을 유지하면서 over blurring을 완화하는 network 제안한다.

Research on Support Vector Machine in Image Denoising

Xinfeng Guo,Chunyan Meng 보안공학연구지원센터 2015 International Journal of Signal Processing, Image Vol.8 No.2

In this paper, a denoising algorithm and simulation experiments of algorithm based on wavelet transform and support vector machine (SVM) image is proposed, a new method is adopted in the selection of characteristic vector of support vector machine, based on training of support vector machine, the support vector machine model is used to distinguish between noise and the original image, to achieve the effect of denoising. The experimental results show that the method can well remove the noise, and can save some important details of images, compared with other denoising method based on wavelet transform, it has a good advantage.