Iterative Back Projection Algorithm Research Articles

Back projection deep unrolling network for handwritten text image super resolution

Current super-resolution (SR) methods have demonstrated exceptional advancements in the domain of natural image processing. Nevertheless, these approaches do not fully address the open issues of edge blurring and distortion. In order to address the aforementioned challenges, we propose a novel deep unrolling model, coined back projection deep unrolling network (BPDUN), for super-resolution of handwritten text images leveraging the Algorithm Unrolling paradigm. We design the network model of BPDUN by unfolding and truncating the traditional iterative back projection algorithm (IBPA). We unroll IBPA into a cascade operation of three building blocks (deep denoising module, low-frequency reconstruction module, and residual projection module). BPDUN inherits the interpretability of the iterative optimization algorithm and is also designed to make the reconstructed text image more realistic and natural. Moreover, we propose a new benchmark dataset to address challenging SR problems of handwritten text image (HDT300). Extensive experiments show that BPDUN obtains an enhanced balance between the performance (quantified by PSNR and SSIM) and the cost (as measured by network parameters). Notably, BPDUN sets new benchmarks on the HDT300 dataset, surpassing previous state-of-the-art approaches by achieving up to 0.2 dB gains in PSNR.

Low light image enhancement using curvelet transform and iterative back projection

With the advancement of technology in image capturing, people are accustomed to high-resolution images. One of the primary necessities of an image capturing system is to provide the same. However, in many cases, the image resolution may not be reaching the expectations of the user which leads to a decrease in user experience. This is a common phenomenon that occurs when the images are captured in low light or if the image encounters a distortion either because of lack of exposure or the image capturing devices may be equipped with a small size sensor. In this work, a resolution enhancement technique using the concepts of curvelet transform and iterative back projection is presented. Sparse representation of images can be enhanced using a combination of curvelet transforms with iterative back projection. Application of curvelet transform along with iterative back projection algorithm on low light images results in enhancing the resolution of the images. The resultant images from here then passed through the inverse transform block and gives an image with contrast enhancement which leads to the user experience improvement. The antiquated image enhancement with improvement in the resolution is validated with the measurement of peak signal-to-noise ratio and structural similarity index. The usage of curvelet transform with iterative back projection leads to the restoration of the image resolution by minimizing the distortions, thus leading to an enhanced image whose edge details are retained.

Improving the iterative back projection estimation through Lorentzian sharp infinite symmetrical filter

<span>This study proposed an enhancement technique for improvising the estimation technique in iterative back projection (IBP) by using the Lorentzian error function with a sharp infinite symmetrical filter (SISEF). The IBP estimation is an iteratively based error correction that can minimize the error reconstruction significantly. However, the IBP has a drawback in that it suffers from jaggy and ringing artifacts as a result of the iterative reconstruction method and the absence of edge guidance. Furthermore, because the IBP estimator tended to oscillate at the same solution frequently, numerous iterations were required. Therefore, this study proposed edge enhancement to enhance the estimator by using the combination of the IBP with Lorentzian SISEF to produce a finer high-resolution output image. As a result, the SISEF is used to improvise the estimator by providing high accuracy of edge detail information for enhancing the edge image. At the same time, the Lorentzian error norm helps to increase the robustness of the IBP algorithm from contamination of additional noise and the ringing artifacts.</span>

Super-Resolution of Intravoxel Incoherent Motion Imaging Based on Multisimilarity

Intravoxel incoherent motion imaging (IVIM) is a new magnetic resonance imaging (MRI) technique that can detect both diffusion and pseudo perfusion microscopic tissue information. However, it requires to acquire the diffusion-weighted (DW) images with multiple b-values, which leads to long acquisition time. To reduce the time, the spatial resolution of IVIM is usually not high enough to reflect the image details. To solve this problem, in this study, a multi-similarity block-based super-resolution reconstruction (SRR) method for IVIM was proposed. The fact that the image block similarity of DW images occurs not only across or within the images at different scales, but also the adjacent slices with the same b-values. The intra-slice and inter-slice similarity at multi-scales is defined as multi-similarity in this work. To fully explore the multi-similarity patches to assist SRR, a chain of Gaussian pyramids of a given image slice and the corresponding adjacent slices with the same b-value are established first, and then a random search strategy is used to find and identify the most similar blocks in low resolution (LR) image pyramids. Next, the matched blocks are mapped to high resolution (HR) image pyramids and then unwrapped by inverse affine transformation to replace original LR blocks. Finally, the super-resolution images are obtained through an iterative back-projection algorithm. To evaluate the performance of proposed method, we compared the brain and liver IVIM reconstruction results with several state-of-art methods. The results indicate that the multi-similarity-based method can achieve the best reconstruction results in both DW images and the corresponding IVIM parameter maps.

Optimizing 4D abdominal MRI: image denoising using an iterative back-projection approach

4D-MRI is a promising tool for organ exploration, target delineation and treatment planning. Intra-scan motion artifacts may be greatly reduced by increasing the imaging frame rate. However, poor signal-to-noise ratios (SNR) are observed when increasing spatial and/or frame number per physiological cycle, in particular in the abdomen.In the current work, the proposed 4D-MRI method favored spatial resolution, frame number, isotropic voxels and large field-of-view (FOV) during MR-acquisition. The consequential SNR penalty in the reconstructed data is addressed retrospectively using an iterative back-projection (IBP) algorithm. Practically, after computing individual spatial 3D deformations present in the images using a deformable image registration (DIR) algorithm, each 3D image is individually enhanced by fusing several successive frames in its local temporal neighborood, these latter being likely to cover common independent informations. A tuning parameter allows one to freely readjust the balance between temporal resolution and precision of the 4D-MRI.The benefit of the method was quantitatively evaluated on the thorax of 6 mice under free breathing using a clinically acceptable duration. Improved 4D cardiac imaging was also shown in the heart of 1 mice. Obtained results are compared to theoretical expectations and discussed. The proposed implementation is easily parallelizable and optimized 4D-MRI could thereby be obtained with a clinically acceptable duration.

Wavelet-based iterative perfect reconstruction in computational integral imaging.

We propose a new computational integral imaging (CII) method via the iterative perfect reconstruction technique to improve the visual quality of reconstructed 3D scenes. As is well known, images reconstructed by CII suffer from artifacts and, as a result, degradation of visual quality. As solutions to this problem, the regularization and iterative back-projection (IBP)-based super-resolution (SR) reconstruction algorithms have been shown to be effective for high-visual-quality reconstruction. However, computation of the regularization algorithm is very expensive, and the IBP algorithm is very sensitive to noise in the deblurring process. To address these challenges, we propose an iterative perfect reconstruction algorithm that addresses the issues of low visual quality and noise sensitivity. Experimental results indicate that our proposed method outperforms the conventional SR reconstruction-based CII methods.

Human Face Super-Resolution Based on Hybrid Algorithm

Aiming at the problems of image super-resolution algorithm with many convolutional neural networks, such as large parameters, large computational complexity and blurred image texture, we propose a new algorithm model. The classical convolutional neural network is improved, the convolution kernel size is adjusted, and the parameters are reduced; the pooling layer is added to reduce the dimension. Reduced computational complexity, increased learning rate, and reduced training time. The iterative back-projection algorithm is combined with the convolutional neural network to create a new algorithm model. The experimental results show that compared with the traditional facial illusion method, the proposed method can obtain better performance.

Single image super-resolution using a deep encoder–decoder symmetrical network with iterative back projection

Image super-resolution (SR) usually refers to reconstructing a high resolution (HR) image from a low resolution (LR) image without losing high frequency details or reducing the image quality. Recently, image SR based on convolutional neural network (SRCNN) was proposed and has received much attention due to its end-to-end mapping simplicity and superior performance. This method, however, only using three convolution layers to learn the mapping from LR to HR, usually converges slowly and leads to the size of output image reducing significantly. To address these issues, in this work, we propose a novel deep encoder-decoder symmetrical neural network (DEDSN) for single image SR. This deep network is fully composed of symmetrical multiple layers of convolution and deconvolution and there is no pooling (down-sampling and up-sampling) operations in the whole network so that image details degradation occurred in traditional convolutional frameworks is prevented. Additionally, in view of the success of the iterative back projection (IBP) algorithm in image SR, we further combine DEDSN with IBP network realization in this work. The new DEDSN-IBP model introduces the down sampling version of the ground truth image and calculates the simulation error as the prior guidance. Experimental results on benchmark data sets demonstrate that the proposed DEDSN model can achieve better performance than SRCNN and the improved DEDSN-IBP outperforms the reported state-of-the-art methods.

Multi-example feature-constrained back-projection method for image super-resolution

Example-based super-resolution algorithms, which predict unknown high-resolution image information using a relationship model learnt from known high- and low-resolution image pairs, have attracted considerable interest in the field of image processing. In this paper, we propose a multi-example feature-constrained back-projection method for image super-resolution. Firstly, we take advantage of a feature-constrained polynomial interpolation method to enlarge the low-resolution image. Next, we consider low-frequency images of different resolutions to provide an example pair. Then, we use adaptive kNN search to find similar patches in the low-resolution image for every image patch in the high-resolution low-frequency image, leading to a regression model between similar patches to be learnt. The learnt model is applied to the low-resolution high-frequency image to produce high-resolution high-frequency information. An iterative back-projection algorithm is used as the final step to determine the final high-resolution image. Experimental results demonstrate that our method improves the visual quality of the high-resolution image.

An improved iterative back projection algorithm based on ringing artifacts suppression

In order to suppress and reduce the ringing artifacts of the IBP super resolution algorithm, an improved iterative back projection (IBP) reconstruction algorithm is presented in this paper. Firstly, two novel concepts, variation and local variance variation based on an adaptive analysis are introduced. They are calculated based on the error projection matrix in the back projection process. Secondly, a restraint item is adopted to constrain the back projection matrix. Thirdly, a modified iterative scheme is used to reconstruct the high resolution image. The effectiveness of the proposed approach on ringing artifacts suppression is demonstrated by experiments. The results indicate that the proposed method can achieve satisfactory performance.

Super resolution in CT

ABSTRACTThe general framework of super resolution in computed tomography (CT) system is introduced. Two data acquisition ways before or after the reconstruction respectively are described. Three models including the sinogram model, the in‐plane model and the z‐axis model, are addressed to adapt super resolution to CT system. The improved iterative back projection algorithm is used in this work. Experimental results based on simulated data, GE performance phantom scanned by GE LightSpeed VCT system, one patient volunteer scanned by TOSHIBA Aquilion system, and a special experimental apparatus demonstrate that super resolution is effective to improve the resolution of CT images. The sinogram model is suitable for future CT system; the in‐plane model is restricted to some special clinical diagnoses; and the z‐axis model is practicable for current general clinical CT images. © 2015 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 25, 92–101, 2015

Video Superresolution Reconstruction Using Iterative Back Projection with Critical-Point Filters Based Image Matching

To improve the spatial resolution of reconstructed images/videos, this paper proposes a Superresolution (SR) reconstruction algorithm based on iterative back projection. In the proposed algorithm, image matching using critical-point filters (CPF) is employed to improve the accuracy of image registration. First, a sliding window is used to segment the video sequence. CPF based image matching is then performed between frames in the window to obtain pixel-level motion fields. Finally, high-resolution (HR) frames are reconstructed based on the motion fields using iterative back projection (IBP) algorithm. The CPF based registration algorithm can adapt to various types of motions in real video scenes. Experimental results demonstrate that, compared to optical flow based image matching with IBP algorithm, subjective quality improvement and an average PSNR score of 0.53 dB improvement are obtained by the proposed algorithm, when applied to video sequence.

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

To improve the spatial resolution of low resolution image with Gaussian blur and Pepper & salt noise, a blind single-image super resolution reconstruction method is proposed. In the low resolution imaging model, the Gaussian blur, down-sampling, as well as Pepper & Salt noise are all considered. Firstly, the Pepper & Salt noise in the low resolution image is reduced through median filtering method. Then, the Gaussian blur of the de-noised image is estimated through error-parameter analysis method. Finally, super resolution reconstruction is carried out through iterative back projection algorithm. Experimental results show that the Gaussian blur is estimated with high accuracy, and the Pepper & Salt noise are removed effectively. The visual effect and peak signal to noise ratio (PSNR) of the super resolution reconstructed image is improved. In addition, the importance of Gaussian blur in single-image super resolution reconstruction is justified in an experimental way.

A New Method for Superresolution Image Reconstruction Based on Surveying Adjustment

A new method for superresolution image reconstruction based on surveying adjustment method is described in this paper. The main idea of such new method is that a sequence of low-resolution images are taken firstly as observations, and then observation equations are established for the superresolution image reconstruction. The gray function of the object surface can be found by using surveying adjustment method from the observation equations. High-resolution pixel value of the corresponding area can be calculated by using the gray function. The results show that the proposed algorithm converges much faster than that of conventional superresolution image reconstruction method. By using the new method, the visual feeling of reconstructed image can be greatly improved compared to that of iterative back projection algorithm, and its peak signal-to-noise ratio can also be improved by nearly 1 dB higher than the projection onto convex sets algorithm. Furthermore, this method can successfully avoid the ill-posed problems in reconstruction process.

Gaussian Noised Single-Image Super Resolution Reconstruction

A framework is proposed to reconstruct a super resolution image from a single low resolution image with Gaussian noise. The degrading processes of Gaussian blur, down-sampling, and Gaussian noise are all considered. For the low resolution image, the Gaussian noise is reduced through Wiener filtering algorithm. For the de-noised low resolution image, iterative back projection algorithm is used to reconstruct a super resolution image. Experiments show that de-noising plays an important part in single-image super resolution reconstruction. In the super reconstructed image, the Gaussian noise is reduced effectively and the peak signal to noise ratio (PSNR) is increased.

Single-Image Super Resolution Reconstruction with Pepper and Salt Noise

In order to improve the resolution of single image with Pepper and Salt noise, a framework is proposed. In the low resolution imaging model, the Gaussian blur, down-sampling, as well as Pepper and Salt noise are considered. For the low resolution image, the Pepper and Salt noise is reduced through median filtering method. Super resolution reconstruction is performed on the de-noised low resolution image by iterative back projection algorithm. Experimental results show that the Pepper and Salt noise are removed effectively and the peak signal to noise ratio (PSNR) of the super resolution reconstructed image is improved.

An Adaptively Edge-guided Back-projection Algorithm for Single Image Super-resolution

The iterative back-projection (IBP) algorithm for single image super-resolution (SR) can not guarantee the edge smoothness because the reconstruction error is back-projected isotropically. Furthermore, in the case that there is only one low-resolution input, the estimation of the reconstruction error is easily affected by noise. Thus, this algorithm is highly sensitive to noise. The focus of this paper is to overcome the two disadvantages. Firstly, we present a novel IBP algorithm which incorporates an adaptively edge-guided interpolation into the back-projection process. Thus, the back-projection process can be guided by the local edge information adaptively. Experimental results show that the proposed algorithm outperforms the state-of-the-art IBP methods. It can generate a highresolution image with smooth edges. Secondly, we propose a robust IBP algorithm which exploits nonlocal redundancies to improve the reliability of the reconstruction error estimation. Experimental results show the proposed algorithm can work well for the noisy images.

Blind Single-Image Super-Resolution Reconstruction Based on Motion Blur

Blind image super-resolution reconstruction is one of the hot and difficult problem in image processing. a framework of blind single-image super-resolution reconstruction algorithm is proposed. In the low-resolution imaging model, the processes of motion blur, down-sampling, and noise are considered. The parameter of motion blur is estimated through an error-parameter analysis method. Utilizing Wiener filtering image restoration algorithm, an error-parameter curve at different motion distance is generated, from which the motion distance of the motion point spread function (PSF) can be estimated approximately. The super-resolution image is reconstructed through the iterative back projection (IBP) algorithm. The experimental results show that motion PSF estimation plays an important role on the quality of the SR reconstructed image, and also demonstrate the effectiveness of the proposed algorithm.

A thin and compact compound-eye imaging system incorporated with an image restoration considering color shift, brightness variation, and defocus

A compact system of the thin observation module by bound optics (TOMBO) imager based on compound-eye imaging has been constructed to demonstrate its advantages over single-eye imaging systems such as thinner hardware. To reconstruct a high-resolution image from low resolution images captured by the compound-eye optics, we propose an image restoration scheme based on the iterative back-projection algorithm with depth map estimated from the disparities on the captured image. The scheme includes suppression of unit-by-unit color shift caused by the offset microlenses and the color filters on the commercial image sensors and deblurring of defocus by geometrical optics using the depth map. In the experiment, three-dimensional objects were captured by the TOMBO imager and reconstructed with the scheme. After the processing, the power spectrum of the captured image was improved by up to 19 dB, and the power spectrum of the effect of the color shift was reduced by 7 dB.

Video superresolution reconstruction based on subpixel registration and iterative back projection

To improve the spatial resolution of video, a superresolution reconstruction method based on a sliding window is proposed utilizing the movement information between frames in the low-resolution video. We propose a registration algorithm based on a four-parameter transformation model through Taylor series expansion, using an iterative solving method as well as the Gaussian pyramid image model to estimate the movement parameters from coarseness to fine. Superresolution frames are reconstructed using an iterative back projection (IBP) algorithm. We also present the suitable length of the sliding window and the reasonable iteration number of the IBP algorithm in the video superresolution reconstruction. Our algorithm is compared to other algorithms on simulated images and actual color videos. Both show that our registration algorithm achieves higher subpixel accuracy than other algorithms, even in the case of large movements, and that the reconstructed video has better visual effects and stronger resolution ability. It can be extensively applied to the superresolution reconstruction of video sequences in which the frames are different from each other mainly by translation and rotation.