Estimation Of Blur Parameters Research Articles

Motion-blurred image restoration framework based on parameter estimation and fuzzy radial basis function neural networks

The restoration of motion-blurred images has always been a complex problem in image restoration. The current single blurred image algorithm cannot very well solve the estimation error of motion blur parameters. A comprehensive motion-blurred image restoration framework is proposed, which includes motion-blurred data generation, blur parameter estimation, and image quality assessment of restored images. First, we designed and used four image data sets with different degrees of blurring. We innovatively propose a blur parameter estimation algorithm based on the particle swarm optimization (B-PSO) algorithm. The Naturalness Image Quality Evaluator (NIQE) is used as the fitness function of the PSO algorithm. The framework also introduces a polynomial-based radial basis function neural network (P-RBFNN) as a new image quality assessment (IQA) method, with good image classification performance. Test results from public datasets show that the proposed framework can accurately estimate blur parameters. The peak signal-to-noise ratio (PSNR) reaches 29.976 dB, the structural similarity (SSIM) reaches 0.9044, and the classification rate is 96%. The proposed restoration framework produces the best image restoration results.

Efficient Image Deblurring Application Using Combination of Blind Deconvolution Method and Blur Parameters Estimation Method

Image deblurring can be formulated as the inverse process of image blurring. The degraded image can be improved and reconstructed to its original form using image deblurring. Image reconstruction in elementary mode begins with estimation of degradation parameters followed with application of some classical restoration approach. This paper introduces a new approach for image restoration. Here, removal of motion blurs is done through blind deconvolution technique. Two motion parameters angle (θ) and length have been considered for estimation. Hough transform method is used for estimation of blur parameters. PSNR and MSE are employed to exploit important image characteristics, such as denoising, error, randomness, etc. Proposed algorithm is tested for accuracy. Experiments are carried out to study important image features, which play a major role during image restoration.

Using a Blur Metric to Estimate Linear Motion Blur Parameters.

Motion blur is a common artifact in image processing, specifically in e-health services, which is caused by the motion of a camera or scene. In linear motion cases, the blur kernel, i.e., the function that simulates the linear motion blur process, depends on the length and direction of blur, called linear motion blur parameters. The estimation of blur parameters is a vital and sensitive stage in the process of reconstructing a sharp version of a motion blurred image, i.e., image deblurring. The estimation of blur parameters can also be used in e-health services. Since medical images may be blurry, this method can be used to estimate the blur parameters and then take an action to enhance the image. In this paper, some methods are proposed for estimating the linear motion blur parameters based on the extraction of features from the given single blurred image. The motion blur direction is estimated using the Radon transform of the spectrum of the blurred image. To estimate the motion blur length, the relation between a blur metric, called NIDCT (Noise-Immune Discrete Cosine Transform-based), and the motion blur length is applied. Experiments performed in this study showed that the NIDCT blur metric and the blur length have a monotonic relation. Indeed, an increase in blur length leads to increase in the blurriness value estimated via the NIDCT blur metric. This relation is applied to estimate the motion blur. The efficiency of the proposed method is demonstrated by performing some quantitative and qualitative experiments.

Non‐uniform image blind deblurring by two‐stage fully convolution network

Deep neural networks have recently demonstrated high performance for deblurring. However, few methods are designed for both non-uniform image blur estimation and removal with highly efficient. In this study, the authors proposed a fully convolutional network that outputs estimated blur and restored image in one feed-forward pass for the non-uniformly blurred image of any input-size. The proposed network contains two subnets. The parameter estimation subnet P-net predicts pixel-wise parameters of multiple blur types with high accuracy. The output of P-net is used as a condition, which guides the blur removal subnet G-net to restore a high quality latent sharp image. P-net and G-net are ultimately integrated into a single framework called PG-net, which guarantees the consistency of parameter estimation and blur removal, thereby improves algorithm efficiency. Experiment results show that the authors blur parameter estimation method as well as their deblurring method outperforms the comparison methods both quantitatively and qualitatively.

A new approach to enhance the estimation of blur parameters in blurred images

One of the main objectives of blur parameters estimation is the restoration of blurred images. A novel blur parameters estimation algorithm is presented in the present paper. For estimating the blur angle of the degraded image, Hough transform is applied upon two parallel linear edges on two sides of the central bright stripe that appear in the frequency domain. For blur length estimation, the proposed method has two sequential phases. In the first phase, the frequency domain is used for feature extraction. Also, for obtaining a superior performance in the algorithm in the second stage, a wavelet-based parameter was used, which can determine the extent of the blur in the image. For the prediction of the blur length in a particular blurred image, RBFNN is employed with frequency-domain features of the related blurred image and blur extent parameter, coming from the wavelet transform as the input. Blur length in blurred images is estimated at a higher accuracy using this method. The concentration of this approach is on motion blur parameter estimation. Through comparison with available approaches, the results of simulations indicate the ability and strength of the suggested approach in predicting blur parameters.

Trigonometry-based motion blur parameter estimation algorithm

Restoration of blurred images requires information about the blurring function, which is generally unknown in practical applications. Identification of blur parameters is essential for yielding blurring function. This paper proposes a technique for estimation of motion blur parameters by formulating trigonometric relationship between the spectral lines of the motion blurred image and the blur parameters. In majority of the existing motion blur parameter estimation approaches, length of motion blur is estimated by rotating the Fourier spectrum to estimated motion angle. This requires angle estimation to be done forehand. The proposed method estimates both, length and angle simultaneously by exploring the trigonometric relation between spectral lines, thereby eliminating the need of spectrum rotation for length estimation. The proposed technique is applied on Berkeley segmentation dataset, Pascal VOC 2007 and USC-SIPI image database. The simulation results prove that the proposed method exhibit better parameter estimation performance as compared to existing state-of-the-art techniques.

Texture Features based Blur Classification in Barcode Images

Blur is an undesirable phenomenon which appears as image degradation. Blur classification is extremely desirable before application of any blur parameters estimation approach in case of blind restoration of barcode image. A novel approach to classify blur in motion, defocus, and co-existence of both blur categories is presented in this paper. The key idea involves statistical features extraction of blur pattern in frequency domain and designing of blur classification system with feed forward neural network.

Identifying Image Composites by Detecting Discrepancies in Defocus and Motion Blur

Image manipulation has become commonplace in today's social context. One of the most common types of image forgeries is image compositing. In recent years, researchers have proposed various methods for detecting such splicing. Most prior approaches to detecting blur post-processing operation suffer from their inability to identify the spliced region when the background region contained nature blur. In this study, we propose a novel algorithm of detecting splicing in blurred images. We use blur parameters estimation through the cepstrum characteristics of blurred images in order to restore the spliced region and the rest of the image. We also develop a new measure to assist in inconsistent region segmentation in restored images that contain large amounts of ringing effect. Experimental results show efficacy of the proposed method even if the images to be tested have been noised with different levels. Compared with other existing algorithms, the proposed method has better robustness against gaussian noise.

基于主成分分析的直线运动模糊参数估计

To estimate the blur parameter of a linear motion blur image accurately and quickly,this paper analyses how the blur length and direction show in a frequency image and a cepstrum image,respectively,and proposes a motion blur parameter estimation method based on the Principal Component analysis(PCA).Firstly,the cepstrum image of the blur image was segmented in a binaryzation based on the Gaussian distribution modeling,and the highlight line region in the cepstrum image was obtained.Then,the principal component of the highlight line was extracted based on the PCA,and the direction of the principal component was the blur direction.After the blur direction was estimated,the Radon transform of frequency image for the blur image under the estimated direction was calculated,then the result of Radon transform was smoothed to reduce some artifacts.Finally,the blur length was estimated via calculating the interval between the two local-minimas of the Radon transform.Experiment results indicate that the errors of the estimated blur direction and length are0.138 4°and 0.273 9 pixel,respectively,and the calculation speed is nearly 10 times faster than that of the traditional estimated method based on Radon method with the same accuracy.It concludes that the proposed method can estimate the blur parameter accurately and rapidly.