• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.4
• /
• pp.61-67
• /
• 2017

Low-dose X-ray fluoroscopic image sequences to avoid radiation exposure risk are contaminated by quantum noise. To restore these noisy sequences, we propose a 3D nonlocal means (NLM) filter based on stochastic distancesed can be applied to the denoising of X-ray fluoroscopic image sequences. The stochastic distance is obtained within motion-compensated noise filtering support to remove the Poisson noise. In this paper, motion-adaptive weight which reflected the frame similarity is proposed to restore the noisy sequences without motion artifact. Experimental results including comparisons with conventional algorithms for real X-ray fluoroscopic image sequences show the proposed algorithm has a good performance in both visual and quantitative criteria.

• Journal of the Institute of Convergence Signal Processing
• /
• v.11 no.4
• /
• pp.270-277
• /
• 2010

The BAMS filter is a kind of wavelet shrinkage filter based on the Bayes estimators with no simulation, therefore it can be used for a real time filter. The denoising efficiency of BAMS filter is seriously affected by the estimated noise variance in each wavelet band. To remove noise in signals in existing BAMS filter, the noise variance is estimated by using the quartile of the finest level of details in the wavelet decomposition, and with this variance, the noise of the level is removed. In this paper, to remove the image noise includingodified quartile of the level of detail is proposed. And by these techniques, the image noises of mid and high frequency bands are removed, and the results showed that the increased PSNR of ab the midband noise, the noise variance estimation method using the monotonic transform and the mout 2[dB] and the effectiveness in denosing of low noise deviation images.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.115-117
• /
• 2021

Currently, as IoT technology develops, monitoring systems are being used in various fields, and image processing is being used in various forms. Image data causes noise due to various causes during the transmission and reception process, and if it is not removed, loss of image information or error propagation occurs. Therefore, denoising images is essential. Typical methods of eliminating Salt and Pepper noise in images include AF, MF, and A-TMF. However, existing methods have the disadvantage of being somewhat inadequate in high-density noise. Therefore, in this paper, we propose an algorithm for determining noise for Salt and Pepper denoising and replacing the central pixel with an original pixel if it is non-noise, and processing the filtering mask by segmenting and averaging it in eight directions. We evaluate the performance by comparing and analyzing the proposed algorithms with existing methods.

• Journal of the Institute of Convergence Signal Processing
• /
• v.5 no.2
• /
• pp.111-118
• /
• 2004

Techniques based on thresholding of wavelet coefficients are gaining popularity for denoising data because of the reasonable performance at the low complexity. The VisuShrink which removes the noise with the universal threshold is one of the techniques. The universal threshold is proportional to the noise deviation and the number of data samples. In general, because the noise deviation is not known, one needs to estimate the deviation for determining the value of the universal threshold. But, only for the finest scale wavelet coefficients, it has been known the way of estimating the noise deviation, so the noise in coarse scales cannot be removed with the VisuShrink. We propose here a new denoising method which removes the noise in each scale except the coarsest scale by Visushrink method. The noise deviation at each band is estimated by the monotonic transform and weighted deviation, the product of estimated noise deviation by the weight, is applied to the universal threshold. By making use of the universal threshold and the Soft-Threshold technique, the noise in each band is removed. The denoising characteristics of the proposed method is compared with that of the traditional VisuShrink and SureShrink method. The result showed that the proposed method is effective in denoising on Gaussian noise and quantization noise.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.3
• /
• pp.321-326
• /
• 2019

Image denoising methods based on block matching are founded on the experimental observations that neighboring patches or blocks in images retain similar features with each other, and have been proved to show superior performance in denoising different kinds of noise. The methods, however, take into account only neighboring blocks in searching for similar blocks, and ignore the characteristic features of the reference block itself. Consequently, denoising performance is negatively affected when outliers of the Gaussian distribution are included in the reference block which is to be denoised. In this paper, we propose an expanded block matching method in which noisy images are first decomposed into a number of bit-planes, then the range of true signals are estimated based on the distribution of pixels on the bit-planes, and finally outliers are replaced by the neighboring pixels belonging to the estimated range. In this way, the advantages of the conventional Gaussian filter can be added to the blocking matching method. We tested the proposed method through extensive experiments with well known test-bed images, and observed that performance gain can be achieved by the proposed method.

• ETRI Journal
• /
• v.41 no.4
• /
• pp.415-425
• /
• 2019

Generative adversarial networks (GAN) have been successfully used in many image restoration tasks, including image denoising, super-resolution, and compression artifact reduction. By fully exploiting its characteristics, state-of-the-art image restoration techniques can be used to generate images with photorealistic details. However, there are many applications that require faithful rather than visually appealing image reconstruction, such as medical imaging, surveillance, and video coding. We found that previous GAN-training methods that used a loss function in the form of a weighted sum of fidelity and adversarial loss fails to reduce fidelity loss. This results in non-negligible degradation of the objective image quality, including peak signal-to-noise ratio. Our approach is to alternate between fidelity and adversarial loss in a way that the minimization of adversarial loss does not deteriorate the fidelity. Experimental results on compression-artifact reduction and super-resolution tasks show that the proposed method can perform faithful and photorealistic image restoration.

• 전자공학회논문지 IE
• /
• v.43 no.4
• /
• pp.65-75
• /
• 2006

This thesis deals with a wavelet-based method of denoising of infrared images contaminated with impulse noise and Gaussian noise, he method of thresholding the wavelet coefficients using derivatives and median absolute deviations of the wavelet coefficients of the detail subbands was proposed to effectively denoise infrared images with noises. Particularly, in order to eliminate the impulse noise the method of generating binary masks indicating locations of the impulse noise was selected. By this method, the threshold values dividing edges and noises were obtained more effectively proving the validity of the denoising method compared with the conventional wavelet shrinkage method.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.9
• /
• pp.128-137
• /
• 2012

When a CCD camera acquires images in the low light level environment, not only the image signals but also noise components are amplified by the AGC (auto gain control) circuit. Since the noise level in the images acquired in the dark is very high, it is difficult to remove noise with existing denoising algorithms that are targeting the images taken in the normal light condition. In this paper, we proposed an adaptive denoising algorithm that can efficiently remove significant noises caused by the low light level. First, the window including a target pixel is transformed to the frequency domain. Then the algorithm compares the characteristics of equally divided four frequency bands. Finally the noises are adaptively removed according to the frequency characteristics. The proposed algorithm successfully improves the quality of low light level images than the existing algorithms do.

• Journal of the Korea Convergence Society
• /
• v.10 no.5
• /
• pp.17-26
• /
• 2019

We aimed to build a patient-based allergy prevention system using the smartphone and focused on the region of interest (ROI) extraction method for Optical Character Recognition (OCR) in the general environment. However, the current ROI extraction method has shown good performance in the experimental environment, but the performance in the real environment was not good due to the noisy background. Therefore, in this paper, we propose the compared methods of reducing noisy background to solve the ROI extraction problem. There five methods used as a SMF, DIN, Denoising Autoencoder(DAE), DAE with Convolution Neural Network(DAECNN) and median filter(MF) with DAECNN (MF+DAECNN). We have shown that our proposed DAECNN and MF+DAECNN methods are 69%, respectively, which is relatively higher than the conventional DAE method 55%. The verification of performance improvement uses MSE, PSNR and SSIM. The system has implemented OpenCV, C++ and Python, including its performance, is tested on real images.

• Korean Journal of Radiology
• /
• v.21 no.10
• /
• pp.1165-1177
• /
• 2020

Objective: To assess the feasibility of applying a deep learning-based denoising technique to coronary CT angiography (CCTA) along with iterative reconstruction for additional noise reduction. Materials and Methods: We retrospectively enrolled 82 consecutive patients (male:female = 60:22; mean age, 67.0 ± 10.8 years) who had undergone both CCTA and invasive coronary artery angiography from March 2017 to June 2018. All included patients underwent CCTA with iterative reconstruction (ADMIRE level 3, Siemens Healthineers). We developed a deep learning based denoising technique (ClariCT.AI, ClariPI), which was based on a modified U-net type convolutional neural net model designed to predict the possible occurrence of low-dose noise in the originals. Denoised images were obtained by subtracting the predicted noise from the originals. Image noise, CT attenuation, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were objectively calculated. The edge rise distance (ERD) was measured as an indicator of image sharpness. Two blinded readers subjectively graded the image quality using a 5-point scale. Diagnostic performance of the CCTA was evaluated based on the presence or absence of significant stenosis (≥ 50% lumen reduction). Results: Objective image qualities (original vs. denoised: image noise, 67.22 ± 25.74 vs. 52.64 ± 27.40; SNR [left main], 21.91 ± 6.38 vs. 30.35 ± 10.46; CNR [left main], 23.24 ± 6.52 vs. 31.93 ± 10.72; all p < 0.001) and subjective image quality (2.45 ± 0.62 vs. 3.65 ± 0.60, p < 0.001) improved significantly in the denoised images. The average ERDs of the denoised images were significantly smaller than those of originals (0.98 ± 0.08 vs. 0.09 ± 0.08, p < 0.001). With regard to diagnostic accuracy, no significant differences were observed among paired comparisons. Conclusion: Application of the deep learning technique along with iterative reconstruction can enhance the noise reduction performance with a significant improvement in objective and subjective image qualities of CCTA images.