Motion Blurred Images Research Articles

Bi-l0-l2-norm regularization for blind motion deblurring

In blind motion deblurring, leading methods today tend towards highly non-convex approximations of the l0-norm, especially in the image regularization term. In this paper, we propose a simple, effective and fast approach for the estimation of the motion blur-kernel, through a bi-l0-l2-norm regularization imposed on both the intermediate sharp image and the blur-kernel. Compared with existing methods, the proposed regularization is shown to be more effective and robust, leading to a more accurate motion blur-kernel and a better final restored image. A fast numerical scheme is deployed for alternatingly computing the sharp image and the blur-kernel, by coupling the operator splitting and augmented Lagrangian methods. Experimental results on both a benchmark image dataset and real-world motion blurred images show that the proposed approach is highly competitive with state-of-the-art methods in both deblurring effectiveness and computational efficiency.

A Unified Optimization Perspective to Single/Multi-observation Blur-Kernel Estimation with Applications to Camera-Shake Deblurring and Nonparametric Blind Super-Resolution

The nonparametric blur-kernel estimation, using either single image or multi-observation, has been intensively studied since Fergus et al.'s influential work (ACM Trans Graph 25:787---794, 2006). However, in the current literature there is always a gap between the two highly relevant problems; that is, single- and multi-shot blind deconvolutions are modeled and solved independently, lacking a unified optimization perspective. In this paper, we attempt to bridge the gap between the two problems and propose a rigorous and unified minimization function for single/multi-shot blur-kernel estimation by coupling the maximum-a-posteriori (MAP) and variational Bayesian (VB) principles. The new function is depicted using a directed graphical model, where the sharp image and the inverse noise variance associated with each shot are treated as random variables, while each blur-kernel, in difference from existing VB methods, is just modeled as a deterministic parameter. Utilizing a universal, three-level hierarchical prior on the latent sharp image and a Gamma hyper-prior on each inverse noise variance, single/multi-shot blur-kernel estimation is uniformly formulated as an $${\varvec{\fancyscript{l}}}_{{0.5}}$$l0.5-norm-regularized negative log-marginal-likelihood minimization problem. By borrowing ideas of expectation-maximization, majorization-minimization, and mean field approximation, as well as iteratively reweighted least squares, all the unknowns of interest, including the sharp image, the blur-kernels, the inverse noise variances, as well as other relevant parameters are estimated automatically. Compared with most single/multi-shot blur-kernel estimation methods, the proposed approach is not only more flexible in processing multiple observations under distinct imaging scenarios due to its independence of the commutative property of convolution but also more adaptive in sparse image modeling while in the meanwhile with much less implementational heuristics. Finally, the proposed blur-kernel estimation method is naturally applied to two low-level vision problems, i.e., camera-shake deblurring and nonparametric blind super-resolution. Experiments on benchmark real-world motion blurred images, simulated multiple-blurred images, as well as both synthetic and realistic low-resolution blurred images are conducted, demonstrating the superiority of the proposed approach to state-of-the-art single/multi-shot camera-shake deblurring and nonparametric blind super-resolution methods.

Sharp image estimation from a depth-involved motion-blurred image

When capturing image of a 3D scene with depth variation, the image blur due to camera shake is no longer spatially invariant but spatially varying which contains the information of the depth of the scene. In this paper, we consider the task of latent image restoration from a single depth-involved motion-blurred (DIMB) image. To easily describe the depth effect, we propose a new framework to model the image blur where the blur kernel is parameterized by the motion curve of the camera and depth value of the scene. Moreover we present an iterative algorithm to alternatively estimate the depth map, motion curve and latent image, based on the minimization of a cost function. To make the estimation of depth map more accurate, a two-stage depth estimation algorithm is also proposed. In the experiment part, the performance of the proposed method is evaluated on many examples including synthetic and real-world cases and shows that our method can produce better results on the DIMB images.

Kinematic model for the space-variant image motion of star sensors under dynamical conditions

A kinematic description of a star spot in the focal plane is presented for star sensors under dynamical conditions, which involves all necessary parameters such as the image motion, velocity, and attitude parameters of the vehicle. Stars at different locations of the focal plane correspond to the slightly different orientation and extent of motion blur, which characterize the space-variant point spread function. Finally, the image motion, the energy distribution, and centroid extraction are numerically investigated using the kinematic model under dynamic conditions. A centroid error of eight successive iterations <0.002 pixel is used as the termination criterion for the Richardson-Lucy deconvolution algorithm. The kinematic model of a star sensor is useful for evaluating the compensation algorithms of motion-blurred images.

BLIND DECONVOLUTION ON UNDERWATER IMAGES FOR GAS BUBBLE MEASUREMENT

Abstract. Marine gas seeps, such as in the Panarea area near Sicily (McGinnis et al., 2011), emit large amounts of methane and carbon-dioxide, greenhouse gases. Better understanding their impact on the climate and the marine environment requires precise measurements of the gas flux. Camera based bubble measurement systems suffer from defocus blur caused by a combination of small depth of field, insufficient lighting and from motion blur due to rapid bubble movement. These adverse conditions are typical for open sea underwater bubble images. As a consequence so called ’bubble boxes’ have been built, which use elaborate setups, specialized cameras and high power illumination. A typical value of light power used is 1000W (Leifer et al., 2003). In this paper we propose the compensation of defocus and motion blur in underwater images by using blind deconvolution techniques. The quality of the images can be greatly improved, which will relax requirements on bubble boxes, reduce their energy consumption and widen their usability.

Restoration of motion-blurred image based on border deformation detection: a traffic sign restoration model.

Due to the rapid development of motor vehicle Driver Assistance Systems (DAS), the safety problems associated with automatic driving have become a hot issue in Intelligent Transportation. The traffic sign is one of the most important tools used to reinforce traffic rules. However, traffic sign image degradation based on computer vision is unavoidable during the vehicle movement process. In order to quickly and accurately recognize traffic signs in motion-blurred images in DAS, a new image restoration algorithm based on border deformation detection in the spatial domain is proposed in this paper. The border of a traffic sign is extracted using color information, and then the width of the border is measured in all directions. According to the width measured and the corresponding direction, both the motion direction and scale of the image can be confirmed, and this information can be used to restore the motion-blurred image. Finally, a gray mean grads (GMG) ratio is presented to evaluate the image restoration quality. Compared to the traditional restoration approach which is based on the blind deconvolution method and Lucy-Richardson method, our method can greatly restore motion blurred images and improve the correct recognition rate. Our experiments show that the proposed method is able to restore traffic sign information accurately and efficiently.

A new approach on restoration of dynamic measurement uncertainties in optical precision coordinate metrology

This paper presents a new approach to the restoration of dynamic influenced measurement uncertainties in optical precision coordinate metrology (OPCM) using image sensors to measure geometrical features. Dynamic measurements within the context of this paper are based upon relative motion between the imaging setup (CCD-camera and optical system) and the measuring object respectively the measuring scene. The dynamic image acquisition causes image motion blur effects, which downgrades the uncertainties of the measurand. The approach presented deals with a new technique to restore motion degraded images using different methods to analyze important image features by extending the famous state of the art Richardson-Lucy image restoration technique using a new convergence criteria based on the variation of the detectable sub-pixel edge position of each iteration.

Automatic detection of motion blur in intravital video microscopy image sequences via directional statistics of log-Gabor energy maps.

Intravital microscopy is an important experimental tool for the study of cellular and molecular mechanisms of the leukocyte-endothelial interactions in the microcirculation of various tissues and in different inflammatory conditions of in vivo specimens. However, due to the limited control over the conditions of the image acquisition, motion blur and artifacts, resulting mainly from the heartbeat and respiratory movements of the in vivo specimen, will very often be present. This problem can significantly undermine the results of either visual or computerized analysis of the acquired video images. Since only a fraction of the total number of images are usually corrupted by severe motion blur, it is necessary to have a procedure to automatically identify such images in the video for either further restoration or removal. This paper proposes a new technique for the detection of motion blur in intravital video microscopy based on directional statistics of local energy maps computed using a bank of 2D log-Gabor filters. Quantitative assessment using both artificially corrupted images and real microscopy data were conducted to test the effectiveness of the proposed method. Results showed an area under the receiver operating characteristic curve (AUC) of 0.95 (AUC = 0.95; 95 % CI 0.93-0.97) when tested on 329 video images visually ranked by four observers.

The Restoration of Motion Blurred Images Based on the Background Modeling

For the mutual effects of camera shake and subject movement, the image generation space varying motion blur. In order to achieve image restoration, firstly dividing the image area using the Gaussian background modeling, and updated model adaptive to improve the speed and convergence accuracy. Then use the total variation (TV) of the L1 model to estimate the regional point spread function (PSF), and adopted the edge density weight to reduce small edge’s interference for the PSF estimates. Eventually to restored image by Wiener filter. Through experimental analysis, compared with other algorithms, our algorithms get better results in the space varies motion-blurred image.

Parameter Estimation and Restoration of Motion Blurred Image

according to the real motion blur image restoration problems, analyze the difference between the image features and Simulation of real blurred images, this paper proposes a method that applied to real image degradation parameter estimation. First calculate the degraded image using cepstrum, taking the cepstrum to binary image using absolute value of minimum gray as the threshold, and then remove the center cross bright line; and then use formula of point to line to calculate the distance of bright fringe direction of binary image, that is direction of motion blur; the direction of motion blur is rotated to the horizontal direction by the degraded image center of rotation axis, divided the autocorrelation method to calculate fuzzy scale. To estimate the point spread function is take into the Wiener filtering algorithm to recover images, image restoration effect prove that parameter estimation results are correct.

Flexible focus function consisting of convex function and image enhancement filter.

We propose a new focus function Λ that, like many of the existing focus functions, consists of a convex function and an image enhancement filter. Λ is rather flexible because for any convex function and image enhancement filter, it is a focus function. We proved that Λ is a focus function using a model and Jensen's inequality. Furthermore, we generated random Λs and experimentally applied them to simulated and real blurred images, finding that 98% and 99% of the random Λs, respectively, have a maximum value at the best-focused image and most of them decrease as the defocus increases. We also applied random Λs to motion-blurred images, blurred images in different-sized windows, and blurred images with different types of noise. We found that Λ can be applied to motion blur and is robust to different-sized windows and different noise types.

A Rapid Restoration Method for Infrared Aero-Optical Degraded Image of High-Speed Reconnaissance Aircraft

In order to meet the requirement of rapidly deblurring aero-optical degraded image of high-speed reconnaissance aircraft, a rapid restoration method is proposed. By using a guide whose shape, size, and position information are unchanged in the reconnaissance images, the degraded function of the motion blurred image is effectively extracted, and then a inverse filter algorithm is employed to restore the blurred image. In our algorithm, all restoration works are finished only in five steps, so it greatly reduces computation time. A motion-blurred image with a resolution of 1600×1200 is restored only in 0.86s. The experimental results show that the proposed method has distinct advantages of fast computation and preferable effect.

Comparison of Motion-blurred Image Restoration Using Wiener Filter and Spatial Difference Technique

Motion blur of image is caused by relative motion between camera and photographed object during protographing. Motion-blurred image restoration is an important part of the image restoration. Image restoration is the process of recovering original image from its degraded version. This paper focuses on image restoration blurred by uniform linear motion. Principle of image restoration technique is introduced. Degradation model and blurred model of horizontal motion are analyzed. Two methods to solve image restoration caused by uniform linear motion are described in detail: Wiener filter in frequency domain and difference method in spatial domain. Comparisons are made between spatial method and the method in frequency domain. Experimental results on test image show that the spatial difference technique has better performance than the method based on Wiener filter.

POSE VERSUS STATE: ARE SENSOR POSITION AND ATTITUDE SUFFICIENT FOR MODERN PHOTOGRAMMETRY AND REMOTE SENSING?

Abstract. We investigate the advantages of using what we call sensor state parameters or sensor state to describe the geometrical relationship between a sensor and the 3D space or the 4D time-space that extend the traditional image pose or orientation (position and attitude) to the image state. In our concept, at some point in time t, the sensor state is a 12-dimensional vector composed of four 3-dimensional subvectors p(t), v(t), γ(t) and ω(t). Roughly speaking, p(t) is the sensor’s position, v(t) its linear velocity, γ(t) its attitude and ω(t) its angular velocity. It is clear that the state concept extends the pose or orientation ones and attempts to describe both a sensor’s statics (p(t), γ(t)) and dynamics (v(t), ω(t)). It is also clear that if p(t), γ(t) are known for all t within some time interval of interest, then v(t) and ω(t) can be derived from the former. We present three methods to compute the state parameters, two for the continuous case and one for the discrete case. The first two methods rely on the availability of inertial measurements and their derived time-Position-Velocity-Attitude (tPVA) trajectories. The first method extends the INS mechanization equations and the second method derives the IMU angular velocities from INS mechanization equations' output data. The third method derives lineal and angular velocities from relative orientation parameters. We illustrate how sensor states can be applied to solve practical problems. For this purpose we have selected three cases: multi-sensor synchronization calibration, correction of image motion blur (translational and rotational) and accurate orientation of images acquired with focal-plane shutters.

Half Blind Restoration Algorithm of Building Image Based on Total Variation

Based on restoration theories, this paper mainly studies the blind restoration method of motion-blurred image, including a joint identification method and a priori fuzzy identification method. For the halfblind restoration problem, a total variation model is provided and a alternate iteration method based on split-Bregman algorithm is proposed to solve this model. By this new model, the parameters in PSF and the real image can be estimated simultaneously. Our method is demonstrated effective with numerical simulations by the restoration of a motion-blurred building image.

Noncausal fractional directional differentiator and blind deconvoluation: motion blur estimation

The problem of blind estimation of motion blur parameters from a single image is addressed. The blur direction and extent of motion-blurred image, which are introduced by relative motion between a camera and its object scene, are needed in the methods of image restoration, such as blind deconvolution. As an extension to the fractional-order derivative, a noncausal fractional-order directional derivative operator is devised, which is robust to noise. Based on this new operator, a novel method identifying blur parameters is developed in this work. The performance comparison between the proposed method and the state-of-the-art method is also presented, demonstrating that the former provides better immunity to noise and capacity to identify motion blur extent, especially for large blur length.

Shape from Sharp and Motion-Blurred Image Pair

Motion blur due to camera shake is a common occurrence. During image capture, the apparent motion of a scene point in the image plane varies according to both camera motion and scene structure. Our objective is to infer the camera motion and the depth map of static scenes using motion blur as a cue. To this end, we use an unblurred---blurred image pair. Initially, we develop a technique to estimate the transformation spread function (TSF) which symbolizes the camera shake. This technique uses blur kernels estimated at different points across the image. Based on the estimated TSF, we recover the complete depth map of the scene within a regularization framework.

Motion blur parameters estimation for image restoration

This paper deals with estimation of parameters for motion blurred images. The objectives are to estimate the length (L) and the blur angle (θ) of the given degraded image as accurately as possible so that the restoration performance can be optimised. Gabor filter is utilized to estimate the blur angle whereas a trained radial basis function neural network (RBFNN) estimates the blur length. Once these parameters are estimated the conventional restoration is performed. To validate the proposed scheme, simulation has been carried out on standard images as well as in real images subjected to different blur angles and lengths. The robustness of the scheme is also validated in noise situations of different strengths. In all situations, the results have been compared with standard schemes. It is in general observed that the proposed scheme outperforms its counterparts in terms of restoration parameters and visual quality.

Research on Image Motion Blur for Low Altitude Remote Sensing

Research on Image Motion Blur for Low Altitude Remote Sensing

Deblurring and Denoising of Magnetic Resonance Images using Blind Deconvolution Method

paper describes the methods to improve the quality of blurred and noisy Magnetic Resonance Images (MRI) due to motion artifact. The main objective of this work is to restore original image from a motion-blurred image due to defocused optical system and motion artifact which is a challenging problem in digital imaging. The blind deconvolution method has been applied to restore the clear image from the blurred and noisy MRI images in this paper. There exist several techniques to restore the original image corrupted by various noise and blurred due to motion artifact, but these methods lack to fetch the originality of image. Blind deconvolution is a method to recover the sharp version of a blurred image when the source of blurring is not known. The aim of this paper is to analyze and evaluate recent blind deconvolution algorithms both theoretically and experimentally for deblurring and noise removal of MRI mges. The MRI data used was corrupted by Gaussian noise and blurred due to various cause (gaussian blurring, out-of-focus blur and motion blur etc). The comparisons of the techniques were done on the basis of Peak signal to noise ratio (PSNR) and Mean structural similarity index (MSSIM). The proposed technique proved to be better over methods used for deblurring and denoising of MRI images.