Accelerating block-matching and 3D filtering method for image denoising on GPUs

Honzátko, David; Kruliš, Martin

doi:10.1007/s11554-017-0737-9

Cited by 31 publications

(31 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As can be seen from Table 1, the proposed method of predicting the λ 3D parameter value based on the input noisy image (instead of using the default value: 2.7) produces better scores with respect to all three metrics (lower MSE, higher PSNR, higher SSIM). It is important to emphasize that, BM3D scores using default parameter values reported in denoising literature are obtained using the traditional CPU-based implementation of BM3D, which is far much slower than the GPU-based real-time implementation we use [17]. From Table 1, the most significant improvement is seen in terms of SSIM score for all noise levels.…”

Section: Analysis and Discussionmentioning

confidence: 98%

“…in photographs captured by consumer cameras [24]. Moreover, efficient GPU implementation of BM3D has significantly improved its time performance [17]. BM3D has a lot of input parameters which need to be tuned, though most published denoising methods (learning and non-learning based) compare their performance with BM3D using its default parameter values, as mentioned in the original BM3D paper [9].…”

Section: Motivationmentioning

confidence: 99%

See 1 more Smart Citation

CNN-Based Real-Time Parameter Tuning for Optimizing Denoising Filter Performance

Mukherjee

Kottayil

Sun

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

We propose a novel direction to improve the denoising quality of filtering-based denoising algorithms in real time by predicting the best filter parameter value using a Convolutional Neural Network (CNN). We take the use case of BM3D, the state-of-the-art filtering-based denoising algorithm, to demonstrate and validate our approach. We propose and train a simple, shallow CNN to predict in real time, the optimum filter parameter value, given the input noisy image. Each training example consists of a noisy input image (training data) and the filter parameter value that produces the best output (training label). Both qualitative and quantitative results using the widely used PSNR and SSIM metrics on the popular BSD68 dataset show that the CNN-guided BM3D outperforms the original, unguided BM3D across different noise levels. Thus, our proposed method is a CNN-based improvement on the original BM3D which uses a fixed, default parameter value for all images.

show abstract

Section: Analysis and Discussionmentioning

confidence: 98%

Section: Motivationmentioning

confidence: 99%

CNN-Based Real-Time Parameter Tuning for Optimizing Denoising Filter Performance

Mukherjee

Kottayil

Sun

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…This is especially true for the block matching‐based QPDIR algorithm. Block matching is essentially a stencil operation and its implementation on GPU has been thoroughly investigated . Considering that (a) block matching is known to scale well with GPU resources and that (b) the bulk of the computational workload for the QPDIR algorithm is dominated by the block matching operation, QPDIR has the potential to be sped up dramatically using state‐or‐the‐art GPU cards or by adopting a multi‐GPU implementation.…”

Section: Discussionmentioning

confidence: 99%

Quadratic penalty method for intensity‐based deformable image registration and 4DCT lung motion recovery

Castillo

2019

Medical Physics

View full text Add to dashboard Cite

Intensity‐based deformable image registration (DIR) requires minimizing an image dissimilarity metric. Imaged anatomy, such as bones and vasculature, as well as the resolution of the digital grid, can often cause discontinuities in the corresponding objective function. Consequently, the application of a gradient‐based optimization algorithm requires a preprocessing image smoothing to ensure the existence of necessary image derivatives. Simple block matching (exhaustive search) methods do not require image derivative approximations, but their general effectiveness is often hindered by erroneous solutions (outliers). Block match methods are therefore often coupled with a statistical outlier detection method to improve results. Purpose The purpose of this work is to present a spatially accurate, intensity‐based DIR optimization formulation that can be solved with a straightforward gradient‐free quadratic penalty algorithm and is suitable for 4D thoracic computed tomography (4DCT) registration. Additionally, a novel regularization strategy based on the well‐known leave‐one‐out robust statistical model cross‐validation method is introduced. Methods The proposed Quadratic Penalty DIR (QPDIR) method minimizes both an image dissimilarity term, which is separable with respect to individual voxel displacements, and a regularization term derived from the classical leave‐one‐out cross‐validation statistical method. The resulting DIR problem lends itself to a quadratic penalty function optimization approach, where each subproblem can be solved by straightforward block coordinate descent iteration. Results The spatial accuracy of the method was assessed using expert‐determined landmarks on ten 4DCT datasets available on http://www.dir-lab.com. The QPDIR algorithm achieved average millimeter spatial errors between 0.69 (0.91) and 1.19 (1.26) on the ten test cases. On all ten 4DCT test cases, the QPDIR method produced spatial accuracies that are superior or equivalent to those produced by current state‐of‐the‐art methods. Moreover, QPDIR achieved accuracies at the resolution of the landmark error assessment (i.e., the interobserver error) on six of the ten cases. Conclusion The QPDIR algorithm is based on a simple quadratic penalty function formulation and a regularization term inspired by leave‐one‐out cross validation. The formulation lends itself to a parallelizable, gradient‐free, block coordinate descent numerical optimization method. Numerical results indicate that the method achieves a high spatial accuracy on 4DCT inhale/exhale phases.

show abstract

“…In the process of IR image SR, the image has low contrast and fuzzy edge. In particular, there are types of denoising algorithms, including variational, Total Variational (TV [25]), Partial Differential Equation (PDE [26,27]), Block-Matching, and the 3D filtering (BM3D [28,29]) method. TV is used to reduce the degradation of flat areas of the image, but it has faults of complex calculations and slow convergence.…”

Section: Ir Image Denoisingmentioning

confidence: 99%

RGB-IR Cross Input and Sub-Pixel Upsampling Network for Infrared Image Super-Resolution

Zhou

Qian

et al. 2020

Sensors

View full text Add to dashboard Cite

Deep learning-based image super-resolution has shown significantly good performance in improving image quality. In this paper, the RGB-IR cross input and sub-pixel upsampling network is proposed to increase the spatial resolution of an Infrared (IR) image by combining it with a color image of higher spatial resolution obtained with a different imaging modality. Specifically, this is accomplished by fusion of the features map of two RGB-IR inputs in the reconstruction of an infrared image. To improve the accuracy of feature extraction, deconvolution is replaced by sub-pixel convolution to upsample image in the network. Then, the guided filter layer is introduced for image denoising of IR images, and it can preserve the image detail. In addition, the experimental dataset, which is collected by us, contains large numbers of RGB images and corresponding IR images with the same scene. Experimental results on our dataset and other datasets demonstrate that the method is superior to existing methods in accuracy and visual improvement.

show abstract

Accelerating block-matching and 3D filtering method for image denoising on GPUs

Cited by 31 publications

References 12 publications

CNN-Based Real-Time Parameter Tuning for Optimizing Denoising Filter Performance

CNN-Based Real-Time Parameter Tuning for Optimizing Denoising Filter Performance

Quadratic penalty method for intensity‐based deformable image registration and 4DCT lung motion recovery

RGB-IR Cross Input and Sub-Pixel Upsampling Network for Infrared Image Super-Resolution

Contact Info

Product

Resources

About