Restoration of blurred images for surface roughness evaluation using machine vision

An attempt is made to evaluate the surface roughness of uniformly moving machined surface (grinding, milling) using machine vision technique. In the case of moving surfaces the images are likely to blur due to the relative motion between the CCD camera and the object to be captured. Hence the degraded image has to be restored by removing distortion due to motion before subsequent analysis. In this work, image blur due to motion is considered, in particular, blur that occurs when the motion is uniform at constant speed and in a fixed direction. The blurred image is modeled as a convolution between the original image and a known point spread function. The Richardson–Lucy Restoration algorithm, a method of estimation based on Bayes theorem has been used to correct the image. The algorithm is tested in simulations and in practical experiments. A simulation gives complete control over the setup and enables to test the performance of the algorithm. The quantification of roughness for restored images are performed using the statistical parameters such as spatial frequency, arithmetic average of gray level and standard deviation after pre-processing. An Artificial Neural Network (ANN) was used with these three statistical parameters as input to predict the vision roughness. Finally, vision roughness values calculated using the deblurred images are compared with the stylus roughness value. An analysis based on the comparison to understand the validity of the present approach of estimation of surface roughness based on the digitally processed images for implementation in practice, is presented in this paper.     

Noise suppression of point spread functions and its influence on deconvolution of three-dimensional fluorescence microscopy image sets

The point spread function (PSF) is of central importance in the image restoration of three-dimensional image sets acquired by an epifluorescent microscope. Even though it is well known that an experimental PSF is typically more accurate than a theoretical one, the noise content of the experimental PSF is often an obstacle to its use in deconvolution algorithms. In this paper we apply a recently introduced noise suppression method to achieve an effective noise reduction in experimental PSFs. We show with both simulated and experimental three-dimensional image sets that a PSF that is smoothed with this method leads to a significant improvement in the performance of deconvolution algorithms, such as the regularized least-squares algorithm and the accelerated Richardson–Lucy algorithm.     

基于光纹特征的激光主动照明图像去模糊

受光学系统离焦、大气扰动、平台振动的影响,激光主动照明系统捕获的图像容易被模糊,而传统的去模糊方法难以取得良好的复原效果,故本文提出基于光纹特征的盲解卷积复原方法来实现图像去模糊。首先将模糊图像降采样,建立尺度金字塔,在尺度空间查找光纹特征图像块。随后基于激光主动照明图像饱和像素较多的特点,提出新的图像退化模型。最后针对模糊核估计、光纹参数更新、清晰图像复原3个步骤,提出适用的能量函数,迭代复原出无噪清晰图像。搭建了主动照明系统,在捕获的激光主动照明图像上进行了实验,并与现有方法进行了对比。结果表明:本文方法不仅能够复原出清晰图像,而且能有效抑制振铃效应,其客观评价指标峰值信噪比(PSNR)优于已有的其他算法。     

航空多重模糊图像的恢复

提出一种针对多重模糊图像的恢复算法,以解决航空成像中多重模糊同时作用时的图像恢复问题.对多重模糊的物理模型进行分析,通过分步去模糊对多重模糊图像进行恢复.对空间不变模糊的点扩散函数进行合并,减少多重模糊恢复过程的计算误差累计,通过一次解卷积运算实现多重空间不变模糊图像的恢复.对多重模糊图像在分步恢复过程中所产生的误差噪声进行分析,使用维纳滤波对阶段误差噪声进行抑制,使得恢复图像的峰值信噪比(PSNR)值提高7.76.实验结果表明基于点扩散函数合并的恢复方法能将多重模糊图像的PSNR值提高到28.09,有效地保障了图像的恢复质量.     

Spatially varying regularization of deconvolution in 3D microscopy

Confocal microscopy has become an essential tool to explore biospecimens in 3D. Confocal microcopy images are still degraded by out‐of‐focus blur and Poisson noise. Many deconvolution methods including the Richardson–Lucy (RL) method, Tikhonov method and split‐gradient (SG) method have been well received. The RL deconvolution method results in enhanced image quality, especially for Poisson noise. Tikhonov deconvolution method improves the RL method by imposing a prior model of spatial regularization, which encourages adjacent voxels to appear similar. The SG method also contains spatial regularization and is capable of incorporating many edge‐preserving priors resulting in improved image quality. The strength of spatial regularization is fixed regardless of spatial location for the Tikhonov and SG method. The Tikhonov and the SG deconvolution methods are improved upon in this study by allowing the strength of spatial regularization to differ for different spatial locations in a given image. The novel method shows improved image quality. The method was tested on phantom data for which ground truth and the point spread function are known. A Kullback–Leibler (KL) divergence value of 0.097 is obtained with applying spatially variable regularization to the SG method, whereas KL value of 0.409 is obtained with the Tikhonov method. In tests on a real data, for which the ground truth is unknown, the reconstructed data show improved noise characteristics while maintaining the important image features such as edges.     

应用半二次罚函数的图像盲去模糊

由于现有的模糊图像盲恢复算法计算复杂度高,计算量大,本文提出了一种基于半二次罚函数的图像盲去模糊算法,并进行了实验验证。应用图像噪声的多阶偏导数的高斯分布特性和图像梯度值服从hyper-Laplacian分布特性建立方程,使用高效交替迭代的算法对方程求解。由于迭代过程中采用快速傅里叶变换一次求解,故大大降低了运算时间,同时获得了很好的恢复效果,为实现实时视频图像去模糊奠定了基础。对一个百万像素级的图像进行了去模糊实验,结果显示,本文算法比当前流行的算法有更快的计算速度和更好的鲁棒性,计算时间缩短了60%。提出的算法为视频图像的实时盲恢复提供了新的工具。     

The influence of the regularization parameter and the first estimate on the performance of Tikhonov regularized non-linear image restoration algorithms

This paper reports studies on the influence of the regularization parameter and the first estimate on the performance of iterative image restoration algorithms. We discuss regularization parameter estimation methods that have been developed for the linear Tikhonov–Miller filter to restore images distorted by additive Gaussian noise. We have performed experiments on synthetic data to show that these methods can be used to determine the regularization parameter of non-linear iterative image restoration algorithms, which we use to restore images contaminated by Poisson noise. We conclude that the generalized cross-validation method is an efficient method to determine a value of the regularization parameter close to the optimal value. We have also derived a method to estimate the regularization parameter of a Tikhonov regularized version of the Richardson–Lucy algorithm.   These iterative image restoration algorithms need a first estimate to start their iteration. An obvious and frequently used choice for the first estimate is the acquired image. However, the restoration algorithm could be sensitive to the noise present in this image, which may hamper the convergence of the algorithm. We have therefore compared various choices of first estimates and tested the convergence of various iterative restoration algorithms. We found that most algorithms converged for most choices, but that smoothed first estimates resulted in a faster convergence.     

基于降质图像-光流场复原的智能轮椅测速

提出一种运动图像去模糊复原和基于仿射运动模型的光流场去抖动方法,以提高智能轮椅中光流里程计测速方法的精度。当轮椅线速度或角速度较大时,导致机载相机成像产生显著的运动模糊;且轮椅机器人的机械抖动也易产生光流场的偏差,进而影响速度估计的精度。针对于此,首先利用一种基于自适应模糊核的运动图像去模糊方法实现图像复原,以改善视频帧质量;其次,针对智能轮椅在行进中的机械抖动,利用随机抽样一致(RANSAC)排异后的光流场,在卡尔曼滤波框架下估计同名像点的仿射运动模型参数,进而实现光流补偿。实验结果表明所提方法能够提升基于光流场的智能轮椅视觉测速精度。     

基于小波变换的正则化盲图像复原算法

提出了一种将小波变换和自适应正则化方法相结合的盲图像复原算法。该算法先对退化后的图像进行小波分解,得到图像在不同子频段的信息;然后针对各个子频段内图像的频率和方向特性,使用不同的自适应正则化复原方法,在图像的低频子频段进行去模糊;高频子频段则进行抑制噪声和保边缘特征;最后通过小波逆变换得到复原后的图像。实验结果表明, MSE减少了1.60,信噪比增量为1.76,算法性能和复原效果相对空间自适应正则化方法,都有一定的提高。     

Signal‐to‐noise ratio enhancement on SEM images using a cubic spline interpolation with Savitzky–Golay filters and weighted least squares error

A new technique based on cubic spline interpolation with Savitzky–Golay smoothing using weighted least squares error filter is enhanced for scanning electron microscope (SEM) images. A diversity of sample images is captured and the performance is found to be better when compared with the moving average and the standard median filters, with respect to eliminating noise. This technique can be implemented efficiently on real‐time SEM images, with all mandatory data for processing obtained from a single image. Noise in images, and particularly in SEM images, are undesirable. A new noise reduction technique, based on cubic spline interpolation with Savitzky–Golay and weighted least squares error method, is developed. We apply the combined technique to single image signal‐to‐noise ratio estimation and noise reduction for SEM imaging system. This autocorrelation‐based technique requires image details to be correlated over a few pixels, whereas the noise is assumed to be uncorrelated from pixel to pixel. The noise component is derived from the difference between the image autocorrelation at zero offset, and the estimation of the corresponding original autocorrelation. In the few test cases involving different images, the efficiency of the developed noise reduction filter is proved to be significantly better than those obtained from the other methods. Noise can be reduced efficiently with appropriate choice of scan rate from real‐time SEM images, without generating corruption or increasing scanning time.     

基于稀疏表示和Weber定律的运动图像盲复原

针对运动过程中视觉图像易产生运动模糊的问题,提出了一种基于稀疏表示和Weber定律相结合的图像盲复原方法。该方法利用冲击滤波器预测模糊图像的显著边缘梯度,并用多尺度策略由粗到细进行模糊核的估计。然后,对图像盲复原模型进行稀疏正则化约束,并结合反映人类视觉特性的Weber定律对合成模糊图像和真实模糊图像进行盲复原。实验结果表明,本文采用的盲复原算法的性能指标和图像的纹理都达到了较优的复原效果。与近年较好的Rob Fergus去模糊方法和Xu Li去模糊方法相比,对Lena模糊图去模糊后的结构相似度(SSIM)为0.762 4,峰值信噪比(PSNR)提高了1.82~2.99dB;对Cameraman模糊图去模糊后的结构相似度(SSIM)为0.8589,PSNR提高了2.46~5.58dB。另外,本文方法降低了复原图像的边界伪影,符合人的视觉感知特性。     

航空多重模糊图像恢复算法

提出一种针对多重模糊的图像恢复算法,解决航空成像中多重模糊同时作用时的图像恢复问题。对多重模糊的物理模型进行分析,通过分步去模糊对多重模糊图像进行恢复。对空间不变模糊的点扩散函数进行合并,减少多重模糊恢复过程的计算误差累计,通过一次解卷积运算实现多重空间不变模糊图像的恢复。对包含空间变换模糊的多重模糊在分步恢复过程中所产生的图像误差噪声进行分析,使用维纳滤波对阶段误差噪声进行抑制,使得恢复图像的PSNR值提高7.76dB。实验结果表明基于点扩散函数合并的恢复方法能将多重模糊图像的PSNR值提高到28.09,有效地保障了图像的恢复质量。     

Parallel deconvolution of large 3D images obtained by confocal laser scanning microscopy

Various deconvolution algorithms are often used for restoration of digital images. Image deconvolution is especially needed for the correction of three‐dimensional images obtained by confocal laser scanning microscopy. Such images suffer from distortions, particularly in the Z dimension. As a result, reliable automatic segmentation of these images may be difficult or even impossible. Effective deconvolution algorithms are memory‐intensive and time‐consuming. In this work, we propose a parallel version of the well‐known Richardson–Lucy deconvolution algorithm developed for a system with distributed memory and implemented with the use of Message Passing Interface (MPI). It enables significantly more rapid deconvolution of two‐dimensional and three‐dimensional images by efficiently splitting the computation across multiple computers. The implementation of this algorithm can be used on professional clusters provided by computing centers as well as on simple networks of ordinary PC machines. Microsc. Res. Tech., 2010. © 2009 Wiley‐Liss, Inc.     

离焦模糊图像的维纳滤波复原研究

微操作中，显微视觉系统获取的图像通常是离焦模糊图像。离焦模糊图像的退化模型可用圆盘函数描述，利用模糊图像无方向性的二阶拉氏微分图像的自相关的负相关峰形成的环形槽的直径等于作为圆盘函数直径的2倍可以确定该函数。对模糊图像进行一次维纳滤波方法得到原图像的估计值，然后利用该初始值求得原图像及噪声的谱密度估值，进而利用这些新获得的信息构成改进的维纳滤波器对退化图像进行第二次滤波。实验表明，该方法计算量小、鉴别精度高、抗噪声能力较强，突出原图像的一些关键细节，提高了图像的复原质量。     

Performance of new signal‐to‐noise ratio estimation for SEM images based on single image noise cross‐correlation

A new technique for estimation of signal‐to‐noise ratio in scanning electron microscope images is reported. The method is based on the image noise cross‐correlation estimation model recently developed. We derive the basic performance limits on a single image signal‐to‐noise ratio estimation using the Cramer–Rao inequality. The results are compared with those from existing estimation methods including the nearest neighbourhood (the simple method), the first order linear interpolator, and the autoregressive based estimator. The comparisons were made using several tests involving different images within the performance bounds. From the results obtained, the efficiency and accuracy of image noise cross‐correlation estimation technique is considerably better than the other three methods.     

Application of the split-gradient method to 3D image deconvolution in fluorescence microscopy

The methods of image deconvolution are important for improving the quality of the detected images in the different modalities of fluorescence microscopy such as wide‐field, confocal, two‐photon excitation and 4Pi. Because deconvolution is an ill‐posed problem, it is, in general, reformulated in a statistical framework such as maximum likelihood or Bayes and reduced to the minimization of a suitable functional, more precisely, to a constrained minimization, because non‐negativity of the solution is an important requirement. Next, iterative methods are designed for approximating such a solution. In this paper, we consider the Bayesian approach based on the assumption that the noise is dominated by photon counting, so the likelihood is of the Poisson‐type, and that the prior is edge‐preserving, as derived from a simple Markov random field model. By considering the negative logarithm of the a posteriori probability distribution, the computation of the maximum a posteriori (MAP) estimate is reduced to the constrained minimization of a functional that is the sum of the Csiszár I‐divergence and a regularization term. For the solution of this problem, we propose an iterative algorithm derived from a general approach known as split‐gradient method (SGM) and based on a suitable decomposition of the gradient of the functional into a negative and positive part. The result is a simple modification of the standard Richardson–Lucy algorithm, very easily implementable and assuring automatically the non‐negativity of the iterates. Next, we apply this method to the particular case of confocal microscopy for investigating the effect of several edge‐preserving priors proposed in the literature using both synthetic and real confocal images. The quality of the restoration is estimated both by computation of the Kullback–Leibler divergence of the restored image from the detected one and by visual inspection. It is observed that the noise artefacts are considerably reduced and desired characteristics (edges and minute features as islets) are retained in the restored images. The algorithm is stable, robust and tolerant at various noise (Poisson) levels. Finally, by remarking that the proposed method is essentially a scaled gradient method, a possible modification of the algorithm is briefly discussed in view of obtaining fast convergence and reduction in computational time.     

一种基于NSCT变换的超声图像降噪新算法

为有效抑制超声仪器成像中固有的斑点噪声,提出了一种基于非降采样Contourlet变换(nonsubsampled Contourlettransform,NSCT)域中边缘信号系数区提取和最小均方误差(minimum mean square error,MMSE)估计的超声图像的降噪算法。根据NSCT变换的细节信息刻画能力和平移不变性,对其各高频子带中系数进行分类,提取出边缘信号和平缓信号系数区;对超声图像的乘性斑点噪声进行推导研究,在边缘信号系数区和平缓信号系数区,根据各自噪声项的性质分别得出满足贝叶斯最小均方误差估计的降噪滤波方程;最后,对降噪后的系数进行NSCT反变换重建得到降噪图像。仿真图像和临床超声图像的实验结果证实,该算法与传统方法相比,不但能更有效地对斑点噪声进行抑制,也更好地保留了图像的细节信息。     

小波变换及其在MEMS谐振器运动轨迹的特征提取中的应用

为测量微电机械系统(MEMS)谐振器的动态特性参数,根据MEMS谐振器运动图像的特点,将小波变换应用于MEMS谐振器运动轨迹的特征提取中.基于模糊图像合成技术,利用小波变换对MEMS谐振器的模糊运动图像进行了增强及降噪处理,并结合传统的图像处理方法,提取MEMS谐振器的运动轨迹,最终获得了MEMS谐振器的特性参数,从而可为MEMS器件的设计提供重要参考.实验结果表明,利用小波变换的方法获得了更好的测量精度,测量重复性误差为100nm.     

Industrial x-ray inspection system with improved image characterization using blind deblurring based on compressed-sensing scheme

An industrial x-ray inspection system has recently established by our group to examine large and dense objects available in industry. It consists of an industrial x-ray generator having a tube voltage of 450?kV and a focal spot size of 1?mm, a flat-panel detector having a pixel size of 200?µm and a pixel dimension of 2048?×?2048, and a mechanical support for object’s installation. For improving the image characteristics of the system, an effective blind deblurring method based on compressed-sensing scheme is reported. Blind deblurring is the image restoration by estimating the original image and the degradation mechanism using partial information on both. Compressed-sensing is a relatively new mathematical theory for solving the inverse problems. Systematic measurements were performed and the image characteristics of the restored images were quantitatively evaluated using several image-quality indicators. The results demonstrate that the deblurring method is effective for industrial x-ray inspection systems.