Blind Single-Image Super Resolution Reconstruction with Gaussian Blur and Pepper &amp; Salt Noise

Qin, Fengqing; Zhu, Lihong; Cao, Lei; Yang, Wei

doi:10.4304/jcp.9.4.896-902

Cited by 7 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Super-resolution is defined as the reconstruction of a high resolution image from multiple low resolution images of the same scene [10]. By using similar images, the details can be used to enhance the image the process is being applied to [11].…”

Section: A Super-resolutionmentioning

confidence: 99%

Medical image segmentation using multi-scale and super-resolution method

Lin

McLaughlin

Alshehri

2014

2014 IEEE Applied Imagery Pattern Recognition Workshop (AIPR)

View full text Add to dashboard Cite

In many medical imaging applications, a clear delineation and segmentation of areas of interest from low resolution images is crucial. It is one of the most difficult and challenging tasks in image processing and directly determines the quality of final result of the image analysis. In preparation for segmentation, we first use preprocessing methods to remove noise and blur and then we use superresolution to produce a high resolution image. Next, we will use wavelets to decompose the image into different sub-band images. In particular, we will use discrete wavelet transformation (DWT) and its enhanced version double density dual discrete tree wavelet transformations (D3-DWT) as they provide better spatial and spectral localization of image representation and have special importance to image processing applications, especially medical imaging. The multi-scale edge information from the sub-bands is then filtered through an iterative process to produce a map displaying extracted features and edges, which is then used to segment homogenous regions. We have applied our algorithm to challenging applications such as gray matter and white matter segmentations in Magnetic Resonance Imaging (MRI) images.

show abstract

Section: A Super-resolutionmentioning

confidence: 99%

Medical image segmentation using multi-scale and super-resolution method

Lin

McLaughlin

Alshehri

2014

2014 IEEE Applied Imagery Pattern Recognition Workshop (AIPR)

View full text Add to dashboard Cite

show abstract

“…The second is based on reconstruction. The HR images are reconstructed as the inverses of the LR images in these methods, such as interpolation [12][13][14][15], Projection onto Convex Sets (POCS) [16][17][18], Maximum a Posteriori [19][20][21][22], Iterative Back Projection [23][24][25], and their combination [26].…”

Section: Introductionmentioning

confidence: 99%

Projection onto Convex Sets Method in Space-frequency Domain for Super Resolution

Shen¹,

Fang²,

Chen³

et al. 2014

JCP

View full text Add to dashboard Cite

The aim of super resolution is to get high resolution (HR) images/videos from low-resolution (LR) images/videos. The obtained HR images/videos are expected to be clear and have less artifacts. Projection Onto Convex Sets (POCS) in the space domain is a super resolution method. It results in edge oscillation and produces many artifacts. This paper introduces a POCS method in both the space and the frequency domain. Firstly, a frequency domain POCS method is proposed. Then it is combined with the space POCS and the space-frequency POCS is obtained. Compared with the common bilinear interpolation method and the existing POCS method in the space domain, our method may decrease the edge oscillation phenomena and raise the Peak Signal to Noise Ratio (PSNR).

show abstract

“…One of the most common phenomena in images is that when an image is captured, there is frequently a blur that prevents the image from being completely clear [1,2]. The blur often comes from transmission and post processing [3], along with how the picture is captured [4], and compressed [5,6]. As a growing field, finding methods for reducing the blur of a digital image has become an interesting challenge for mathematicians and computer scientists of recent years.…”

Section: Introductionmentioning

confidence: 99%

Modified deconvolution using wavelet image fusion

McLaughlin

Lin

Blasch

et al. 2014

2014 IEEE Applied Imagery Pattern Recognition Workshop (AIPR)

View full text Add to dashboard Cite

Image quality is affected by two predominant factors, noise and blur. Blur typically manifests itself as a smoothing of edges, and can be described as the convolution of an image with an unknown blur kernel. The inverse of convolution is deconvolution, a difficult process even in the absence of noise, which aims to recover the true image. Removing blur from an image has two stages: identifying or approximating the blur kernel, then performing a deconvolution of the estimated kernel and blurred image. Blur removal is often an iterative process, with successive approximations of the kernel leading to optimal results. However, it is unlikely that a given image is blurred uniformly. In real world situations most images are already blurred due to object motion or camera motion/defocus. Deconvolution, a computationally expensive process, will sharpen blurred regions, but can also degrade the regions previously unaffected by blur. To remedy the limitations of blur deconvolution, we propose a novel, modified deconvolution, using wavelet image fusion (moDuWIF), to remove blur from a no-reference image. First, we estimate the blur kernel, and then we perform a deconvolution. Finally, wavelet techniques are implemented to fuse the blurred and deblurred images. The details in the blurred image that are lost by deconvolution are recovered, and the sharpened features in the deblurred image are retained. The proposed technique is evaluated using several metrics and compared to standard approaches. Our results show that this approach has potential applications to many fields, including: medical imaging, topography, and computer vision.

show abstract

Blind Single-Image Super Resolution Reconstruction with Gaussian Blur and Pepper & Salt Noise

Cited by 7 publications

References 14 publications

Medical image segmentation using multi-scale and super-resolution method

Medical image segmentation using multi-scale and super-resolution method

Projection onto Convex Sets Method in Space-frequency Domain for Super Resolution

Modified deconvolution using wavelet image fusion

Contact Info

Product

Resources

About