一种补偿平移与旋转运动的快速电子稳像算法

针对视频图像序列的非稳特性,研究了电子稳像算法中灰度投影算法和块匹配算法各自的不足之处,提出了一种快速补偿视频图像序列间平移及旋转运动的稳定成像算法.该算法先采用灰度投影算法估计并补偿视频图像序列间的平移运动,再利用拉普拉斯变换在靠近图像的边缘区域选取几个具有明显特征的小块,运用块匹配算法进行匹配,计算并补偿其旋转运动量,以得到稳定的视频图像序列.通过理论分析和实验验证,表明这种稳像算法具有速度快、准确度高的特点.     

最小核值相似区低层次图像处理算法的改进及应用

首先介绍一种能有效地进行边缘、角点检测和滤波等低层次图像处理的最小核值相似区算法,然后提出自适应阈值的选取方法,局部区域灰度重心判据对其算法的改进使得边缘检测算法抗噪能力更强。针对序列图像的具体应用,用改进的边缘检测算法能准确、快速地从噪声图像中得到较准确的边缘信息,用滤波算法对序列图像作预处理,可使互相关跟踪结果更可靠、更准确。     

最小核植相似区低层次图像处理算法的改进及应用

首先介绍一种能有效地进行边缘、角点检测和滤波等低层次图像处理的最小核植相似区算法,然后提出自适应阈值的选取方法,局部区域灰度重心判据对其算法的改进使得边缘检测算法抗噪能力更强。针对序列图像的具体应用,用改进的边缘检测算法能准确、快速地从噪声图像中得到较准确的边缘信息,用滤波算法对序列图像作预处理,可使互相关跟踪结果更可靠。     

基于回形窗的弱小多目标图像分割方法

提出一种红外弱小多目标图像分割方法,用一个回形窗口和对比度阈值分割图像.对天空背景下低信噪比的红外弱小多目标图像序列能够有效的分割,抑制噪音干扰.将该方法与传统的图像分割方法做了比较,并对用不同阈值,不同窗口分割时的分割结果进行了分析.实验表明,该算法在执行效率和检测概率上能够取得满意的结果.     

基于Contrast box算法的图像序列有用目标段搜寻方法

提出一种基于Contrast box算法的红外弱小多目标图像序列有用目标段搜寻方法,引用每帧图像的信号量来搜索红外图像序列中含有目标的有用信息段.描述了Contrast box算法和搜索有用目标段处理过程,分析了阈值权值系数和目标所占像元面积选取的不同对图像有用目标段搜索的影响,给出了每幅图像的统计信号量.实验表明,该方法在执行效率和搜索时间上能够取得满意的结果.     

基于人眼视觉的对不良照明图像的二值化方法

提出一种新的基于视觉特性的自适应阈值分割方法.利用人眼视觉对比敏感度特性把图像块分成两类,分别借助OTSU算法和模拟人眼识别过程的多尺度模糊隶属方法实现对这两类具有不同灰度特性图像块的自动阈值分割.实验结果表明,使用该方法能够有效的克服不良照明的影响,还原图像的原本特征信息,二值化效果较好.     

应用于多视点视频编码的基于深度特征的图像区域分割与区域视差估计

基于多视点视图深度特征,提出一种通过简单块匹配运算划分多视点视图区域并估计区域视差的算法.首先基于深度对象的概念确定图像中具有不同深度的区域数量以及这些区域对应的区域视差,再根据误差最小化准则初步确定每个图像块所属区域.当区域中图像块数量小于某个阈值时,采用区域合并算法将该区域中的每个图像块合并到与它的视差最为接近的其它图像区域,通过迭代形成最终的有效图像区域划分.实验表明,该算法能够以图像块为基本单元有效地划分各深度层区域,并准确估计对应的区域视差.     

一种基于时域滤波的红外序列图像去噪算法

红外图像是现代光学设备常用的图像源,图像显示效果直接影响设备的用户体验,而红外序列图像中的噪声会导致显示效果的下降。为了减轻噪声对红外序列图像显示效果的影响,通过历史多帧灰度值的加权和当前帧无噪声图像的估计,提出了一种基于时域高斯滤波的去噪方法。参考空域双边滤波的权值分配方法,引入了灰度值的影响对时域高斯滤波的权值进行修正,解决时域滤波导致的序列图像中运动目标拖尾和模糊。实验结果表明,时域滤波方法能够有效平滑帧间噪声,减轻噪声导致的红外序列图像显示效果的恶化,引入灰度值的影响进行滤波权值修正之后,能够解决时域滤波导致的运动目标拖尾和模糊问题。     

基于多特征和FCM的图像边缘检测方法

提出了一种新的基于多特征和FCM的边缘检测算法.该方法根据边缘点附近灰度分布特点构造了多个反映边缘特性的特征分量,并利用输入图像提取该组特征分量,组成一个反映图像边缘特征的数据集.用FCM聚类算法将该数据集分为两类,即边缘点数据和非边缘点数据,实现边缘检测.该方法无需确定阈值,对弱边缘检测较敏感,在特征的选取上充分考虑了边缘和噪声的本质区别,因而具有优异的抗噪性能.     

一种多像素图像边缘提取方法

提出了一种基于相邻像素间的灰度差异来提取图像多像素边缘的方法.对一幅256×256大小的图像进行边缘提取需要的时间约0.22 s,分别比Prewitt算子和Robert算子少了26%和34%.算法改进后,较完整地提取了图像的竖直边缘和水平边缘,这些边缘在Prewitt算子和Robert算子的提取结果中是不可见的.经细化算法后,得到了图像清晰、位置精确的单像素边缘图像.研究表明,本文的多像素边缘提取等诸方法具有模型简单、实时性强等特点,且便于实现.     

动态测量的高光谱图像压缩感知

针对高光谱图像相邻波段之间具有强光谱相关性的特点,为了提高高光谱图像压缩感知的重构效果,本文提出一种利用边缘信息设计动态测量率的压缩感知算法。首先,通过随机投影的分块压缩感知方法对每个图像块以固定测量率采样,重构出单波段图像作为其他波段的先验信息,并对其提取出图像边缘区域;然后,根据每个图像块边缘信息的丰富程度来自适应分配测量值。在固定总测量数的前提下,对不同图像块分配不同的测量次数。最后,利用分配好的测量次数对其余波段进行采集和重构。仿真结果表明,在相同总测量数情况下,本文提出的动态测量算法重构出的高光谱图像质量（PSNR）与传统固定测量压缩感知策略相比提高了1~4 dB,相比较下的重构时间也减少,在成功重构高光谱图像的基础上更增强了细节处的图像质量。     

基于小波域三状态HMT模型的含噪图像增强

针对含噪图像增强问题,提出一种基于小波域三状态隐马尔可夫树模型的方法,采用三状态的高斯混合模型逼近小波系数的分布,不需要设定精确的阈值,依据期望最大算法训练得到的每个系数所属状态的后验概率,将系数区分为噪声系数、弱边缘系数和强边缘系数,然后通过抑制噪声系数,增强细节特征系数来达到对含噪图像增强的目的,并引入循环平移策略避免人工失真.通过对含噪的标准图像和人脑核磁共振图像进行仿真实验,并与几种经典的图像增强方法作视觉上的对比和定量分析.实验结果表明,本文所提出的方法具有很好的鲁棒性,在突出了图像中更多的细节信息的同时,可以有效抑制噪声.     

An adaptive integrated rule-based algorithm for license plate localization

This paper addresses a license plate localization (LPL) algorithm for a complex background. Most of LPL algorithm works on restricted conditions, as well as on a principle of sequential elimination of blocks from image level to final LP candidate region. In most of algorithms, blocks are filtered out for not satisfying required LP features in a top-down approach and this may result in a poor efficiency in a complex scenario. The major steps of the proposed approach are adaptive edge mapping, saliency measure of edge based rules with confidence level estimation using fuzzy rules and final step for reassessment of decision by colour attributes filtering. The proposed algorithm is adaptive to across the country variations in LP standards, as well as it is tested on two data sets each one consisting of more than 700 images, set-1 being for good images while set-2 including only constrained images. The algorithm is tested for a low contrast due to overexposure or poor lighting, existence of multiple plates, variation in aspect ratio and compatible background conditions. It has been observed, that the performance degradation imposing complex condition is nominal.     

基于人工鱼群优化分块的多聚焦图像融合

在多聚焦图像的融合过程中,对源图像采用固定大小的分块会导致融合后的图像存在块效应、边缘模糊甚至聚焦错误。为了克服此问题,提出了一种新的基于人工鱼群优化分块的多聚焦图像融合方法。首先,将源图像分解成互不重叠的方块,利用聚焦准则选取清晰度高的方块,将已选择的方块合并重构成初始融合图像。然后,利用改进的人工鱼群优化算法,根据一定的适应度值,寻找最优大小的分块方式,获得更优的融合图像。该方法与基于空域、频域及其他优化算法的融合方法进行了多个实验比较,结果表明,该方法获得的融合图像具有较好的客观质量和主观视觉感觉。     

Vessel boundary extraction based on a global and local deformable physical model with variable stiffness

Reliable and efficient vessel cross-sectional boundary extraction is very important for many medical magnetic resonance (MR) image studies. General purpose edge detection algorithms often fail for medical MR images processing due to fuzzy boundaries, inconsistent image contrast, missing edge features, and the complicated background of MR images. In this regard, we present a vessel cross-sectional boundary extraction algorithm based on a global and local deformable model with variable stiffness. With the global model, the algorithm can handle relatively large vessel position shifts and size changes. The local deformation with variable stiffness parameters enable the model to stay right on edge points at the location where edge features are strong and at the same time, fit a smooth contour at the location where edge features are missing. Directional gradient information is used to help the model to pick correct edge segments. The algorithm was used to process MR cine phase-contrast images of the aorta from 20 volunteers (over 500 images) with excellent results.     

一种电磁干扰源成像多分辨率分区算法

在利用抛物反射面对电磁干扰源成像过程中,由于系统衍射受限导致干扰源成像模糊,分辨率低,难以分辨,由于不同频率不同区域干扰源所成图像分辨率不同,具有分区域多分辨率的特征,采用已有超分辨算法难以提高分辨率。利用Mean Shift算法,在原有算法基础上改进使其能够适应多分辨率的电磁干扰源成像,在图像分割的基础上对多分辨率图像进行分块抽离,并采用基于L_R迭代的盲反卷积算法分别对各区域进行分辨率的提高,仿真结果表明算法能够适应对干扰源的多分辨率电磁成像并提高分辨率。     

Variational infrared image enhancement based on adaptive dual-threshold gradient field equalization

Infrared images are characterized by low signal to noise ratio (SNR) and fuzzy texture edges. This article introduces the variational infrared image enhancement algorithm based on gradient field equalization with adaptive dual thresholds. Firstly, we transform the image into gradient domain and get the gradient histogram. Then, we do the gradient histogram equalization. By setting adaptive dual thresholds to qualify the gradients, the image is prevented from over enhancement. The total variation (TV) model is adopted in the reconstruction of the enhanced image to suppress noise. It is shown from experimental results that the image edge details are significantly enhanced, and therefore the algorithm is qualified for enhancement of infrared images in different applications.     

Interlaced scan CCD image motion deblur for space-variant motion blurs

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.     

A computer program for automated step edge motion analysis from scanning probe microscopy images

A computer algorithm was developed to automatically track the displacement of straight step edges between sequential scanning probe microscopy images of single-crystal surfaces. The program utilizes the Canny edge detection algorithm followed by the Hough Transform of the edge map to identify step edges according to their direction, relative to the image axes, and according to their displacement, relative to the image origin. The tracking of individual steps is facilitated by the fact that straight edges in general maintain their direction and therefore, steps of similar displacement but different direction can be sorted. The algorithm is based on the assumption that the rate of image acquisition is much greater than the rate of (mono)layer growth/dissolution, requiring that changes in step displacement are small in successive images. The change in step displacement in sequential images leads directly to the calculation of the step speed. By tabulating all changes in step displacement through a sequence of images, a statistical representation of the step edge data is produced. The program was evaluated using a sequence of 20 atomic force microscopy images from a calcite (104) surface growing from a supersaturated aqueous solution. The program required, in total, 5 CPU-minutes running on a Pentium 4 processor to compute the mean step speed with 60% precision whereas the equivalent number of measurements performed “by hand” required 6 person-hours at 70% precision. For comparable output, the computer program therefore represents a factor of about 100 decrease in required effort.     

改进次最佳检测在侧扫声呐底混响抑制中的应用*

侧扫声呐进行沉底小目标探测时,底混响是主要背景干扰。底混响通常是一种非平稳、非高斯的带限噪声,它使得白噪声条件下的滤波器性能受到限制。在混响背景下常利用自回归模型对接收信号进预行白化处理,但对于实际侧扫声呐应用,白化后直接匹配滤波的处理效果不甚理想。针对此问题,在自回归模型预白化的基础上,提出采用一种次最佳检测与多分辨二分奇异值分解相结合的改进方法。该方法首先对接收信号进行分段处理,利用改进Burg算法估计每段数据自回归模型的系数及阶数;然后构造白化滤波器对分段数据预白化,并对白化后的数据进行多分辨二分奇异值分解;最后应用ostu方法对原始声图和处理后的声图进行目标检测。仿真与实验结果表明,该方法明显提高了信混比,改善了侧扫声呐沉底静态小目标的成图质量,有利于后期实现基于图像的目标自动检测。