Deblurring Technique Research Articles

Investigation of blur kernel of terahertz images

The paper discusses issues of digital processing of terahertz images. It is shown that despite the improvement of the hardware part of imaging setups, the acquired images still often have a low resolution and suffer from noise and blurring effects. Thus, to improve their visual quality, it is advisable to use special digital processing methods. While some progress has already been made in terms of denoising of terahertz images, the research of their deblurring is only at the very early stage. Therefore, this paper attempts to analyze the properties of blur functions for real terahertz images to further use them while designing a corresponding deblurring technique. For this purpose, the phase-only image method has been used. A study of blur properties for the three most common blur types (defocus, motion and Gaussian blur) has shown that for test images they can be distinguished and their main parameters can be assessed. However, the application of this method to real terahertz images has shown that the blur characteristics in them are very different from the ones obtained for modelled examples. The real blur demonstrates a quite complex behaviour and estimating its kernel requires additional research.

Chest tomosynthesis deblurring using CNN with deconvolution layer for vertebrae segmentation.

Limited scan angles cause severe distortions and artifacts in reconstructed tomosynthesis images when the Feldkamp-Davis-Kress (FDK) algorithm is used for the purpose, which degrades clinical diagnostic performance. These blurring artifacts are fatal in chest tomosynthesis images because precise vertebrae segmentation is crucial for various diagnostic analyses, such as early diagnosis, surgical planning, and injury detection. Moreover, because most spinal pathologies are related to vertebral conditions, the development of methods for accurate and objective vertebrae segmentation in medical images is an important and challenging researcharea. The existing point-spread-function-(PSF)-based deblurring methods use the same PSF in all sub-volumes without considering the spatially varying property of tomosynthesis images. This increases the PSF estimation error, thus further degrading the deblurring performance. However, the proposed method estimates the PSF more accurately by using sub-CNNs that contain a deconvolution layer for each sub-system, which improves the deblurringperformance. To minimize the effect of the spatially varying property, the proposed deblurring network architecture comprises four modules: (1) block division module, (2) partial PSF module, (3) deblurring block module, and (4) assembling block module. We compared the proposed DL-based method with the FDK algorithm, total-variation iterative reconstruction with GP-BB (TV-IR), 3D U-Net, FBPConvNet, and two-phase deblurring method. To investigate the deblurring performance of the proposed method, we evaluated its vertebrae segmentation performance by comparing the pixel accuracy (PA), intersection-over-union (IoU), and F-score values of reference images to those of the deblurred images. Also, pixel-based evaluations of the reference and deblurred images were performed by comparing their root mean squared error (RMSE) and visual information fidelity (VIF) values. In addition, 2D analysis of the deblurred images were performed by artifact spread function (ASF) and full width half maximum (FWHM) of the ASFcurve. The proposed method was able to recover the original structure significantly, thereby further improving the image quality. The proposed method yielded the best deblurring performance in terms of vertebrae segmentation and similarity. The IoU, F-score, and VIF values of the chest tomosynthesis images reconstructed using the proposed SV method were 53.5%, 28.7%, and 63.2% higher, respectively, than those of the images reconstructed using the FDK method, and the RMSE value was 80.3% lower. These quantitative results indicate that the proposed method can effectively restore both the vertebrae and the surrounding softtissue. We proposed a chest tomosynthesis deblurring technique for vertebrae segmentation by considering the spatially varying property of tomosynthesis systems. The results of quantitative evaluations indicated that the vertebrae segmentation performance of the proposed method was better than those of the existing deblurringmethods.

Breaking down Polyblur: Fast Blind Correction of Small Anisotropic Blurs

Polyblur is a two stage blind deblurring technique for removing small-sized blurs, like small camera shake or the lens point-spread function, proposed in 2021 by Delbracio et al. First, the blur is modeled with a zero-mean anisotropic Gaussian kernel whose parameters are rapidly estimated from the oriented blurry image gradients. Second, a sharp estimate is obtained by applying an approximate deconvolution filter, which is designed as a polynomial function of the estimated blurring kernel. Since in practice true blurs are not exactly Gaussian filters, the residual blur is gradually removed by repeating this two-stage procedure. Because it relies only on simple image manipulations, Polyblur is a quick blind deblurring technique, running in a fraction of a second on a smartphone. In this presentation, we analyze its key ingredients, showcase several use cases on real images, and provide Numpy and Pytorch implementations.

Improvement and Assessment of a Blind Image Deblurring Algorithm Based on Independent Component Analysis

The aim of the present paper is to improve an existing blind image deblurring algorithm, based on an independent component learning paradigm, by manifold calculus. The original technique is based on an independent component analysis algorithm applied to a set of pseudo-images obtained by Gabor-filtering a blurred image and is based on an adapt-and-project paradigm. A comparison between the original technique and the improved method shows that independent component learning on the unit hypersphere by a Riemannian-gradient algorithm outperforms the adapt-and-project strategy. A comprehensive set of numerical tests evidenced the strengths and weaknesses of the discussed deblurring technique.

Image De-Blurring Based on Constraint Conditional Model

Image capturing is more vulnerable to the various physical limitations such as defocus, low lighting and camera shaking; this makes the image blurry and noisy. Moreover De-blurring is the process to recover the original image from the given degraded image. De-blurring technique uses the estimated blur kernel for achieving the optimal restored image with the sharp features, however the accuracy has been one of the major concern , hence in this paper we use Constrained Conditional model (CCM) for restoring the image. Moreover, here two different methods are integrated i.e. conditional model and convergence operator, these two combined learns the model and efficiently and provides the better results. In order to evaluate the proposed model, Levin dataset is used by considering the two eminent model metric i.e. PSNR and SSIM and CCM based model outperforms the other state-of-art technique.

Image De-Blurring Using the Least Square Filter Technique in A Neural Network

Image restoration is one of the digital image processing techniques used to repaint or restore image information. One common problem encountered during restoration is image blurring. To solve this problem, a new de-blurring technique was proposed to reduce or remove image blurring. The proposed filter was designed using the least square interpolation calculation controlled by the neural network to select the blurred pixels. The wavelet decomposition technique was used to improve the performance of the proposed filter, which showed good results for fully and partially blurred regions in images.

An Efficient Image De-Blurring Technique Using Point Spread Function in High Definition Medical Image

An Efficient Image De-Blurring Technique Using Point Spread Function in High Definition Medical Image

Application of image quality assessment module to motion-blurred wood images for wood species identification system

Despite tighter conservation regulations, demand for timber products has continued to increase due to growing population. Normally, experts identify the wood species based on the pattern of the wood surface texture. However, manual inspection on wood texture is tedious, time-consuming, impractical and cost-ineffective for a human to analyze a large number of timber species. Therefore, a reliable automatic wood recognition system is needed in order to classify the wood species efficiently. The proposed system includes image acquisition, image quality assessment module (IQA), image deblurring, feature extraction and classification. In this research, the wood images are motion-blurred due to imperfections in the imaging and capturing process. Hence, an IQA module is proposed to monitor the quality of images before proceeding to the next stage which is the feature extraction process. The IQA module will determine whether the image has to undergo the image deblurring process based on the image quality value. If the image is of low quality based on the image quality value obtained, then the image will be deblurred before the feature extraction procedure. A reliable motion deblurring technique, which is based on Lucy–Richardson algorithm, is employed to enhance the motion-blurred images before proceeding to the next stage, which is the feature extraction process. Then, a statistical feature extraction technique is proposed to extract 24 features from each wood image. Finally, a support vector machine is used to classify the 20 tropical wood species.

Temporal-Spatial Consistency and Sampling Neural Network Based Adaptive Hand-held Video Deblurring

Video deblurring for hand-held camera is vital in many high-level video enhancement applications. Although quite a few proposed approaches have achieved remarkable success in recent years, many technical challenges still prevail for video (from hand-held camera) containing wide-range scenes and highly-variable objects. In particular, we are lacking effective and versatile strategies to adaptively handle high-level alignment, parallax diminution, unclear area detection, sharpness enhancement in a consistent fashion. To ameliorate, this paper develops a novel method to successively respect temporal-spatial alignment and precise deblurring. Our aim is to devise a new adaptive video deblurring technique by resorting to new modeling strategies. This paper's key originality is hinged upon the joint utility of both self-adaptive intrinsic mode functions (IMFs) based on empirical mode decomposition (EMD) in the temporal domain for video signal and mesh-structure constraint enforcement in the spatial domain, beside, this paper innovatively jointly utilities the subsampling convolution, upsampling convolution and luckiness threshold value based back propagation, which takes video deblurring as a high-precision adaptive convolution. As a result, our new approach can adaptively improve the clarity of the video, and synchronously optimize the camera trajectory of wobbly video. To validate our optimization approach for adaptive video deblurring, we conduct comprehensive experiments on public benchmarks, and make extensive and quantitative evaluations with available state-of-the-art methods as well as popular commercial software. All of our experiments demonstrate the advantages of the optimization method in terms of versatility, accuracy, and efficiency.

Detection and restoration of multi-directional motion blurred objects

In this paper, an object blur detection and deblurring technique is proposed to restore multi-directional motion blurred objects in a single image. We have proposed local blur angle detection method based on Radon transform (RT) and Laplacian of Gaussian (LoG). While capturing the images, motion blur occurs mainly due to either movement of the objects or movement of the camera. Here, we have focused to restore the objects which has been blurred by motion of the objects. The estimation of likely blur direction is calculated in the blurred image using RT and gradient operators. To detect blur angle locally at each pixel, the new local blur angle estimator using RT and LoG has been developed. Numerical experiments have been carried out for the proposed method, and the results are compared with the state-of-the-art methods.

Evaluation of a novel elastic respiratory motion correction algorithm on quantification and image quality in abdomino-thoracic PET/CT.

Our aim is to evaluate in phantom and patient studies a recently developed elastic motion debluring (EMDB) technique which makes use of all the acquired PET data and compare its performance to other conventional techniques such as phase based gating (PBG) and HDChest (HDC) both of which use fractions of the acquired data. Comparisons were made with respect to static whole-body (SWB) images with no motion correction. Methods: A phantom simulating respiratory motion of the thorax with lung lesions (5 spheres with ID=10- 28 mm) was scanned with 0, 1, 2, and 3 cm motion. Four reconstructions were performed: SWB, PBG, HDC, and EMDB. For PBG, the average (PBGave) and maximum bin (PBGmax) were used. To compare the reconstructions, the ratios of SUVmax (RSmax), SUVpeak (RSpeak), and CNR (RCNR) were calculated with respect to SWB. Additionally, 46 patients with lung or liver tumors < 3 cm diameter were also studied. Measurements of SUVmax, SUVpeak, and contrast-to-noise ratio (CNR) were made for 46 lung and 19 liver lesions. To evaluate image noise, the SUV standard deviation was measured in healthy lung and liver tissue and in the phantom background. Finally, subjective image quality of patient exams was scored on a 5 point scale by four radiologists. Results: In the phantom, EMDB increased SUVmax/SUVpeak over SWB but to a lesser extent than the other reconstruction methodologies. The RCNR for EMDB however was higher than all other reconstructions (0.68 EMDB > 0.54 HDC > 0.41 PBGmax > 0.31 PBGave). Similar results were seen in patient studies. The SUVmax/SUVpeak were higher by 19.3/11.1% EMDB, 21.6/13.9% HDC, 22.8/12.8% PBGave, and 45.6/26.8% PBGmax compared to SWB. Lung/liver noise increased EMDB (3/15%), HDC (35/56%), PBGave (100/170%), and PBGmax (146/219%). CNR increased in lung/liver tumors only for EMDB (18/13%), and decreased for HDC (-14/-23%), PBGave (-39/-63%), and PBGmax (-18/-46%). Average radiologist scores of image quality were SWB (4.0 ± 0.8) > EMDB (3.7 ± 1.0) > HDC (3.1 ± 1.0) > PBG (1.5 ± 0.7). Conclusion: The EMDB algorithm had the least increase in image noise, improved lesion CNR, and had the highest overall image quality score.

Improved field-mapping and artifact correction in multispectral imaging.

To develop a method for improved B0 field-map estimation, deblurring, and image combination for multispectral imaging near metal. A goodness-of-fit field-map estimation technique is proposed that uses only the multispectral imaging (MSI) data to estimate the field map. Using the improved field map, a novel deblurring technique is proposed that also employs a new image combination scheme to reduce the effects of noise and other residual MSI artifacts. The proposed field-map estimation and deblurring techniques are compared to the current methods in phantoms and/or in vivo from subjects with knee, hip, and spinal metallic implants. Phantom experiments validate that the goodness-of-fit field-map estimation is less sensitive to noise and bias than the conventional center-of-mass technique, which reduces distortion in the deblurring methods. The new deblurring approach also is substantially less sensitive to noise and distortion than the current deblurring method, as demonstrated in phantoms and in vivo, and is able to find a good tradeoff between deblurring and distortion. The proposed methods not only enable field-mapping with reduced noise sensitivity but are able to create deblurred images with less distortion and better signal-to-noise ratio with no additional scan time, thereby enabling improved visualization of underlying anatomy near metallic implants. Magn Reson Med 78:2022-2034, 2017. © 2017 International Society for Magnetic Resonance in Medicine.

Double-Tip Artefact Removal from Atomic Force Microscopy Images.

The Atomic Force Microscope (AFM) allows the measurement of interactions at interfaces with nanoscale resolution. Imperfections in the shape of the tip often lead to the presence of imaging artefacts such as the blurring and repetition of objects within images. Generally, these artefacts can only be avoided by discarding data and replacing the probe. Under certain circumstances (e.g., rare, high value samples, or extensive chemical/physical tip modification) such an approach is not feasible. Here, we apply a novel deblurring technique, using a Bayesian framework, to yield a reliable estimation of the real surface topography without any prior knowledge of the tip geometry (blind reconstruction). A key contribution is to leverage the significant recently successful body of work in natural image deblurring to solve this problem. We focus specifically on the 'double-tip' effect, where two asperities 1 are present on the tip, each contributing to the image formation mechanism. Finally, we demonstrate that the proposed technique successfully removes the 'double-tip' effect from high resolution AFM images which demonstrate this artefact whilst preserving feature resolution.

Unsupervised spatial-resolution enhancement of electron beam measurement using deconvolution

This paper presents an unsupervised de-blurring technique that enhances the spatial resolution of electron beam (EB) diagnostic instruments. The inherent degradation in EB diagnosis waveforms is modeled using the convolution between the EB current distribution and the sensor characteristic function. Due to the use of various sensors, the diagnosis results are device dependent. Sensor inaccuracy has a detrimental effect on the deconvolution-based restoration of the ground truth signal. In this research this adverse effect is mitigated by an accurate sensor's Point Spread Function (PSF) derivation, which allows the successful restoration of waveforms. As approximate size of the sensor is known, a probability distribution is assigned to the expected interval of PSF features in the frequency domain, which increases the accuracy of the PSF analysis. By deploying this method, sensor inaccuracies are considered in the dynamic PSF formation, hence providing a promising consistency to the restoration. Restoration is performed with Wiener inverse filter and blind-deconvolution and results are compared with the ground truth pulses, obtained using a sensitive sensor (18um diameter). Experimental results confirm that the purposed method delivers a universal device independent measurement, facilitates instrument production, and EB characterization.

Estimating accurate optical flow in the presence of motion blur

Spatially varying motion blur in video results from the relative motion of a camera and the scene. How to estimate accurate optical flow in the presence of spatially varying motion blur has received little attention so far. We extend the classical warping-based variational optical flow method to deal with this issue. First, we modify the data term by matching the identified nonuniform motion blur between the input images according to a fast blur detection and deblurring technique. Importantly, a downsample-interpolation technique is proposed to improve the blur detection efficiency, which saves 75% or more running time. Second, we improve the edge-preserving regularization term at blurry motion boundaries to reduce boundary errors that are caused by blur. The proposed method is evaluated on both synthetic and real sequences, and yields improved overall performance compared to the state-of-the-art in handling motion blur. © 2015 SPIE and IS&T (DOI: 10.1117/1.JEI.24.5.053018)

Performance Enhancement of Digital ImageDeblurring Using Hybrid Approach

Motion blur can significantly degrade the quality of an image. The images become blurred due to camera movements, object movements or displacements of pixels. Deblurring of an image is done by processing an image through curvelet transform and removal of noise by median filter. Experimental results show that proposed deblurring technique and accuracy measurement using peak signal to noise ratio, root mean square error and mean square error. Curvelet transformation is a multi-scale transformation technique which is most suitable for the objects with curves. Curvelet are a non-adaptive technique for multi scale object representation

Laplacian Erosion: An Image Deblurring Technique for Multi-Plane Gamma-Cameras

Laplacian Erosion, an image deblurring technique for multi-plane Gamma-cameras, has been developed and tested for planar imaging using a GEANT4 Monte Carlo model of the Pixelated Emission Detector for RadioisOtopes (PEDRO) as a test platform. A contrast and Derenzo-like phantom composed of 125I were both employed to investigate the dependence of detection plane and pinhole geometry on the performance of Laplacian Erosion. Three different pinhole geometries were tested. It was found that, for the test system, the performance of Laplacian Erosion was inversely proportional to the detection plane offset, and directly proportional to the pinhole diameter. All tested pinhole geometries saw a reduction in the level of image blurring associated with the pinhole geometry. However, the reduction in image blurring came at the cost of signal to noise ratio in the image. The application of Laplacian Erosion was shown to reduce the level of image blurring associated with pinhole geometry and improve recovered image quality in multi-plane Gamma-cameras for the targeted radiotracer 125I.

An Interactive Deblurring Technique for Motion Blur

interactive deblurring technique to restore a motion blurred image is proposed in this paper. Segment based semi- automated restoration method is proposed using an error gradient descent iterative algorithm. In this approach, segments are automatically detected which are the best representatives of motion blur. Then the decimal parameters of the blur kernel are interactively derived; with extended precision using interpolation between pixels, with comparatively much lower error convergence rate. Once blur kernel is obtained, image is restored using Striling's interpolation formula. Experimental results show that proposed method gives sufficient restoration as interactive judgment gives the most desirable quality.

Enhanced resolution for EPR imaging by two-step deblurring

The broad spectrum of spin probes used for electron paramagnetic resonance imaging (EPRI) result in poor spatial resolution of the reconstructed images. Conventional deconvolution procedures can enhance the resolution to some extent but obtaining high resolution EPR images is still a challenge. In this work, we have implemented and analyzed the performance of a postacquisition deblurring technique to enhance the spatial resolution of the EPR images. The technique consists of two steps; noniterative deconvolution followed by iterative deconvolution of the acquired projections which are then projected back using filtered backprojection (FBP) to reconstruct a high resolution image. Further, we have proposed an analogous technique for iterative reconstruction algorithms such as multiplicative simultaneous iterative reconstruction technique (MSIRT) which can be a method of choice for many applications. The performance of the suggested deblurring approach is evaluated using computer simulations and EPRI experiments. Results suggest that the proposed procedure is superior to the standard FBP and standard iterative reconstruction algorithms in terms of mean-square-error (MSE), spatial resolution, and visual judgment. Although the procedure is described for 2D imaging, it can be readily extended to 3D imaging.

Deblurring of breathing motion artifacts in thoracic PET images by deconvolution methods

In FDG-PET imaging of thoracic tumors, blurring due to breathing motion often significantly degrades the quality of the observed image, which then obscures the tumor boundary. We demonstrate a deblurring technique that combines patient-specific motion estimates of tissue trajectories with image deconvolution techniques, thereby partially eliminating breathing-motion induced artifacts. Two data sets were used to evaluate the methodology including mobile phantoms and clinical images. The clinical images consist of PET/CT co-registered images of patients diagnosed with lung cancer. A breathing motion model was used to locally estimate the location-dependent tissue location probability function (TLP) due to breathing. The deconvolution process is carried by an expectation-maximization (EM) iterative algorithm using the motion-based TLP. Several methods were used to improve the robustness of the deblurring process by mitigating noise amplification and compensating for motion estimate uncertainties. The mobile phantom study with controlled settings demonstrated significant reduction in underestimation error of concentration in high activity case without significant superiority between the different applied methods. In case of medium activity concentration (moderate noise levels), less improvement was reported (10%-15% reduction in underestimation error relative to 15%-20% reduction in high concentration). Residual denoising using wavelets offered the best performance for this case. In the clinical data case, the image spatial resolution was significantly improved, especially in the direction of greatest motion (cranio-caudal). The EM algorithm converged within 15 and 5 iterations in the large and small tumor cases, respectively. A compromise between a figure-of-merit and entropy minimization was suggested as a stopping criterion. Regularization techniques such as wavelets and Bayesian methods provided further refinement by suppressing noise amplification. Our initial results show that the proposed method provides a feasible framework for improving PET thoracic images, without the need for gated/4-D PET imaging, when 4-D CT is available to estimate tumor motion.