首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到19条相似文献,搜索用时 187 毫秒
1.
一种混合模糊含噪图像的参数辨识方法   总被引:1,自引:0,他引:1  
在航空摄影、景象匹配导航等领域,由光学相机采集的图像往往同时存在着运动模糊和散焦模糊。成像中的噪声进一步加大了混合模糊图像点扩展函数的参数辨识难度。提出了一种双谱混合模糊含噪图像的点扩展函数参数辨识方法。首先计算出经混合模糊后的标准测试图像所对应的运动和散焦模糊的双谱,然后通过曲线拟合出双谱中各自的统计特性与运动模糊尺度或散焦半径之间的函数关系,由此训练出的BP神经网络可以完成对其它含有噪声的混合模糊图像点扩展函数的参数辨识。实验结果表明,该方法适用于含有噪声的一定模糊参数范围内的混合模糊图像。在信噪比为25dB的情况下,辨别出的模糊参数偏差不超过0.5个像素。  相似文献   

2.
为了提高汉语语音的谎言检测准确率,提出了一种对信号倒谱参数进行稀疏分解的方法。首先,采用小波包滤波器组对语音信号进行多频带划分,求得子频带对数能量并进行离散余弦变换以提取小波包频带倒谱系数,结合梅尔频率谱系数得到倒谱参数;其次,依据K-奇异值分解方法分别利用说谎和非说谎两种状态下的语音倒谱参数集训练得到过完备混合字典,在此字典上根据正交匹配追踪算法对参数集进行稀疏编码提取稀疏特征;最终进行多种分类模型下的识别实验·实验结果表明,稀疏分解方法相比传统参数降维方法具有更好的优化性能,本文推荐的稀疏谱特征最佳识别率达到78.34%,优于其他特征参数,显著提高了谎言检测识别准确率。   相似文献   

3.
在使用相机对空中远处点目标成像时,不可避免的,会因为相机的角运动引起图像的模糊。针对这一问题,提出了一种运动模糊复原方法。首先通过建立点目标运动的模型来了解造成运动模糊的原因,其次通过对模型的分析,从基本理论开始推导得出一个退化函数,并且采用模糊图像的维纳滤波复原模型。最后,对图像进行了仿真实验。实验结果表明,该复原模型有较好的复原效果。  相似文献   

4.
恒星识别以及卫星目标检测识别是空间监视系统的重要应用之一。由于星图图像点目标成像的特点以及大量背景恒星的干扰,星图中用于目标识别的特征难以提取,因此目标的位置是实现目标识别的关键特征。高斯曲面拟合方法是使用较为广泛的目标质心提取算法之一,通过理论分析和实验表明传统高斯曲面拟合方法对运动卫星的定位存在较大误差。为此,提出了各向异性的高斯曲面拟合模型,该模型通过使用两个不同的高斯模糊参数和旋转因子,可以捕捉目标不同方向的各异特征,适合卫星由于运动造成的随机方向模糊。仿真实验和真实数据实验表明,本文方法的总体定位精度可分别达到0.008和0.04,并能够准确提取星图目标的质心,相比传统方法有较大改善。  相似文献   

5.
提出了一种文本无关说话人识别的全特征矢量集模型及互信息评估方法,该模型通过对一组说话人语音数据在特征空间进行聚类而形成,全面地反映了说话人语音的个性特征。对于说话人语音的似然度计算与判决,则提出了一种互信息评估方法,该算法综合分析距离空间和信息空间的似然度,并运用最大互信息判决准则进行识别判决。实验分析了线性预测倒谱系数(LPCC)和Mel频率倒谱系数(MFCC)两种情况下应用全特征矢量集模型和互信息评估算法的说话人识别性能,并与高斯混合模型进行了比较。结果表明:全特征矢量集模型和互信息评估算法能够充分反映说话人语音特征,并能够有效评估说话人语音特征相似程度,具有很好的识别性能,是有效的。  相似文献   

6.
运动模糊图像恢复的核心是点扩散函数的估计和直接去卷积算法。针对快速运动而形成的低信噪比和小模糊长度图像模糊的问题,提出了一种新的算法来估计模糊核函数的参数,在确定模糊核函数后,模糊图像的恢复采用了一种自然图像梯度统计先验的直接解卷积算法,实验结果证明,与R.Fergus的算法相比较,对于线性运动造成的图像模糊有更快的速度和更好的恢复效果。  相似文献   

7.
实现运动模糊图像恢复的关键是获得准确的点扩散函数(point spread function,PSF)。针对PSF的模糊尺度和模糊方向这两个重要参数,提出了一种基于图像频谱全局均值标准差的估计方法。通过研究运动模糊图像的产生机理,分析了二次傅里叶变换的频谱特性,利用全局均值标准差法对频谱图进行阈值分割,结合中心点和端点的坐标距离来估计模糊尺度,运用Radon变换鉴别模糊方向,实现了PSF参数的确定。进行了标准图像阈值分割准确度的验证,利用仿真模糊图像估计算法进行了验证,并进行了实际复原实验。结果表明,该方法具有误差小、稳定性高和不受运动方向限制的特点,为图像恢复提供了一种准确、简便的参数估计方法。  相似文献   

8.
孙辉  李志强 《中国光学》2012,5(2):174-180
为估算匀速直线运动模糊图像的运动参数,提出了一种基于相位相关分析的图像配准方法。该方法利用傅里叶变换的平移特性,对产生平移的目标图像进行傅里叶变换并计算归一化互功率谱,其傅里叶逆变换对应二维脉冲函数,通过计算脉冲函数峰值坐标获取位移图像之间的亚像元级位移量。结合相位相关配准原理和线性空间不变退化模型,给出了匀速直线运动点扩散函数及其光学传递函数的数学描述;讨论了匀速直线运动模糊对相位相关配准结果的影响,证明了图像经过匀速直线运动退化后,位移图像之间归一化互功率谱具有不变性。实验结果表明:动态运动模糊图像最大检测误差为0.489 pixel,标准差为0.16 pixel。  相似文献   

9.
李永乐  张茂军  娄静涛  王炜 《光学学报》2012,32(9):911001-88
随着高分辨率传感器和大光圈的采用,光圈和反射面曲率造成的折反射全向成像散焦模糊问题越发突出。提出了一种有效去除散焦模糊的折反射全向成像系统设计。理论分析折反射成像散焦模糊的原因,建立全向图点扩展函数与实景空间物点及成像系统虚像位置的关系;在一次曝光成像时间内匀速旋转镜头对焦环,通过累积曝光使全向图散焦模糊核具有期望的空间不变性;利用反卷积算法对散焦模糊全向图进行复原,得到全局清晰的全向图像。该方法较好地解决了折反射全向成像散焦模糊问题,对提高折反射全向成像质量,促进其在相关领域的广泛应用具有重要意义。  相似文献   

10.
高频振动模糊图像仿真与分析   总被引:1,自引:0,他引:1  
研究了运动模糊图像在空间域和频域上的数学模型,根据运动图像的模糊原理,提出了时间等分运动模糊图像的仿真方法,仿真了高频振动模糊图像.理论分析指出,物体做高频谐波振动时运动模糊成像的频谱会由于Bessel函数的零点的存在而形成系列条纹,对振动模糊图像的频谱做两次Radon变换能检测出振动的方向和振幅,从而得到高频振动模糊...  相似文献   

11.
In the problem of blind image deconvolution, estimation of blurring kernel is the first and foremost important step. Quality of restored image highly depends upon the accuracy of this estimation. In this paper we propose a modified cepstrum domain approach combined with bit-plane slicing method to estimate uniform motion blur parameters, which improves the accuracy without any manual intervention. A single motion blurred image under spatial invariance condition is considered. It is noted that the fourth bit plane of the modified cepstrum carries an important cue for estimating the blur direction. With the exploration of this bit plane no other post processing is required to estimate blur direction. The experimental evaluation is carried out on both real-blurred photographs and synthetically blurred standard test images such as Berkeley segmentation dataset and USC-SIPI texture image database. The experimental results show that the proposed method is capable of estimating blur parameters more accurately than the existing methods.  相似文献   

12.
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.  相似文献   

13.
离焦模糊图像超分辨力盲复原算法分析   总被引:2,自引:0,他引:2  
根据计算机模拟和实际拍摄的离焦模糊标准分辨力靶标图像,对提出的离焦模糊图像超分辨力盲复原算法进行了分析研究,对比了超分辨力盲复原算法与几种典型复原算法的效果,以及在不同离焦模糊半径下该算法的复原效果。实验结果表明,该算法对深度离焦模糊图像具有较明显的复原效果,复原图像相比较于模糊图像分辨力提高1倍;显著改善了标准分辨力靶标图像的可分辨组块。可分辨组块可以定量地描述复原算法对图像分辨力的改善程度。该算法已经应用到实际工作当中,并获得了较好的效果。  相似文献   

14.
Recovery of degraded images due to motion blurring is a challenging problem in digital imaging. Most existing techniques on blind deblurring are not capable of removing complex motion blurring from the blurred images of complex structures. One promising approach is to recover the clear image using multiple images captured for the scene. However, in practice it is observed that such a multi-frame approach can recover a high-quality clear image of the scene only after multiple blurred image frames are accurately aligned during pre-processing, which is a very challenging task even with user interactions. In this paper, by exploring the sparsity of the motion blur kernel and the clear image under certain domains, we propose an alternative iteration approach to simultaneously identify the blur kernels of given blurred images and restore a clear image. Our proposed approach is not only robust to image formation noises, but is also robust to the alignment errors among multiple images. A modified version of linearized Bregman iteration is then developed to efficiently solve the resulting minimization problem. Experiments show that our proposed algorithm is capable of accurately estimating the blur kernels of complex camera motions with minimal requirements on the accuracy of image alignment. As a result, our method is capable of automatically recovering a high-quality clear image from multiple blurred images.  相似文献   

15.
A different approach, aiming to achieve the constant blur status of point-spread function (PSF) at a certain defocused plane, is described. The correlation between the two PSF is used to control the PSF blur similarity, and simultaneously the Strehl ratio is also used to control the PSF blur minimization. By designing the PSF so that it is significantly insensitive to defocus or related defocus quantity, for example, due to temperature change, all the constantly blurred images can be accurately de-blurred by a simple inverse restoration filter for an adequate range of defocus. We refer to that as “software lens compensation” and apply a design method to solve the athermalization of middle wavelength infrared (MWIR) imaging systems. The resultant PSF is almost invariant in the temperature range from −10 to 50°C at the same focal plane. Consequently, the constant blur spot can be removed by a simple digital signal processing. Thus, clear and sharp de-blurred images at different temperatures are obtained.  相似文献   

16.
Motion is a common phenomenon in the real world so the consistence of virtual and real motion is one of hot focuses in augmented reality system. This paper mainly analyzes virtual and real motion blur consistence in augmented reality system, and mainly do improvement in two aspects, one is how to get the motion blur parameters, the other is how to render motion blur. From the available literature it can be seen that, most achieved methods to get motion blur parameters are based on machine vision. But this paper first acquires the motion trajectory by magnetic tracking, and then gets motion blur parameters. Considering the available literature on motion blur rendering methods, they mainly contain convolution and accumulator caching and so on. Due to rotation translation compound motion blur, this paper does rendering in advance based on improved mathematical model, and then render virtual object through looking up table in augmented reality system. Experimental results show that the new virtual and real motion blur consistence render algorithm has low time complexity and achieve the desired effect in simulation environment and real scenes.  相似文献   

17.
基于微分图像自相关的离焦模糊图像盲复原   总被引:7,自引:1,他引:6  
赵琳  金伟其  陈翼男  苏秉华 《光学学报》2008,29(9):1703-1709
针对离焦模糊图像的盲复原算法的研究具有重要的实际意义和实用价值.根据光学离焦成像模型,研究提出了一种基于微分图像自相关的离焦模糊图像超分辨力盲复原算法,即首先采用拉普拉斯算子对离焦模糊图像进行二阶微分并求微分图像的自相关,然后从自相关结果所包含的信息中确定离焦模糊半径,最后以离焦模糊模型结合MPMAP超分辨力复原算法对离焦模糊图像进行肓复原.实验证明:算法能够以较高的精度估计出离焦模糊半径并实现离焦模糊图像的盲复原,该算法较其它同类算法在减少计算过程中需要考虑的各类因素的同时也减少了计算量,提高了结果精度,依靠超分辨力复原算法获取更多的复原图像信息,已在实际刑侦和物证鉴定的离焦模糊图像判读和鉴定中获得成功应用.  相似文献   

18.
In this study, the effect of defocus blur on the avoidance of interocular suppression in half-occluded zones, which are the monocular zones formed by occlusion, was examined. Six human adults observed a pair of colored random-dot stereograms, and the cumulative time of disappearance of the half-occluded zones was measured. There were three defocus blur conditions (blurry-background, blurry-occluder, and nonblur) and two fixation conditions (fixation-front and fixation-back). The author found that the cumulative time of disappearance was shorter under the blurrybackground condition of the fixation-front condition and under the blurry-occluder condition of the fixation-back condition than under the other conditions. These results indicate that ecologically valid defocus blur contributes to the avoidance of interocular suppression in the half-occluded zones.  相似文献   

19.
Ting-Fa Xu  Peng Zhao 《Optik》2011,122(8):719-723
Motion blur is caused by camera shakes or object motions during exposure when the shutter speed is relatively slow. As for the object motion blur, the degradation of a CCD image is often characterized by space-variant motion blurs, since objects are often moving in different directions at different speeds. But most image restorations for space-variant motion blurs are addressed only for progressive scan CCD images. To address the space-variant image restorations for interlaced scan images, we propose a novel image restoration scheme. First, one interlaced scan image frame is required, which is divided into the odd field and the even field images. These two field images are further segmented into rectangular blocks. The motion vectors are computed in these rectangular blocks using an efficient block matching algorithm. Second, image restoration is performed in these rectangular blocks using a constrained least square algorithm in the odd or even field image, which can both preserve edge structures and remove noises. Our novel scheme is illustrated by restoring a space-variant blurred moving boat image and a synthetic blurred image.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号