车牌定位及倾斜矫正方法研究

为提高车牌定位的有效性和准确性,利用车牌的纹理特征,提出基于边缘检测和形态学运算相结合的车牌定位算法。在摄取车牌图像的过程中,会因为各种原因导致车牌图像有不同程度的倾斜变形,而倾斜的车牌很难被分割和识别,必须进行倾斜矫正,使用了基于Hough变换的倾斜矫正算法。实验结果表明该车牌定位算法和倾斜矫正算法快速有效。     

一种基于模板库的车牌字符识别算法

模板匹配法在汽车牌照自动识别系统中已经得到广泛应用。应用此方法的系统有一个问题是,后期车牌字符识别过程中对图像处理花费的时间、数据运算量很大。为解决此问题,文中在传统的模板匹配方法的基础上提出了一种新的基于模板库的的模式匹配改进型算法。通过建立模板库,规避了传统算法中车牌的倾斜矫正这一环节,减少了系统运行时的数据运算量,提高了系统处理速度。增强其在车牌倾斜时的识别率、准确率同时,还有很好的鲁棒性。     

基于Hough变换的PDF417条码图像倾斜矫正

在对PDF417条码图像扫描输入的过程中,条码图像不可避免地发生倾斜,因此必需先对图像进行倾斜矫正,才能实现条码的准确识别.Hough变换具有抗噪声能力强的特点,是一种检测图像倾斜角度的重要方法.为了克服Hough变换计算量大的缺点,先通过行差运算提取图像水平边缘,然后只对提取的边缘进行两级Hough变换,实现倾斜角度的检测.本文采用基于直线拟合的方法实现倾斜图像的矫正.与传统的Hough变换以及其它的倾斜图像矫正方法相比,本文实现了准确性与快速性的很好结合.     

基于先验信息和射影几何变换的车牌图像矫正*

为了便于字符分割、识别,车牌识别系统需要将车牌的倾斜图像矫正为无倾斜和旋转的标准图像。目前多数文献采用Hough变换检测倾斜角度并直接进行矫正,但存在的缺陷。为此从射影几何的观点提出了新算法：利用车牌固有的先验信息,将车牌与车牌图像之间的透视变换矫正为仿射变换,再将仿射变换矫正为相似变换,将此相似变换看做是对原车牌进行旋转和全局缩放的结果,消除其中的旋转因素,就得到一个与车牌仅差一个缩放因子的标准车牌图像,以便于后续的字符处理。     

一种基于模板库的车牌字符识别算法

模板匹配法在汽车牌照自动识别系统中已经得到广泛应用.应用此方法的系统有一个问题是,后期车牌字符识别过程中对图像处理花费的时间、数据运算量很大.为解决此问题,文中在传统的模板匹配方法的基础上提出了一种新的基于模板库的的模式匹配改进型算法.通过建立模板库,规避了传统算法中车牌的倾斜矫正这一环节,减少了系统运行时的数据运算量,提高了系统处理速度.增强其在车牌倾斜时的识别率、准确率同时,还有很好的鲁棒性.     

车牌识别系统

在机动车牌照牌识别系统的设计中，车牌区域的检测和牌照中字符的分割是进行字符识别前必须的两个步骤，实验证明，利用车牌的纹理特征和形状特征检测车牌区域具有较高的准确性，算法的实现以边缘检测技术和数学形态学为基础，字符分割受车牌倾斜角度的影响较大，在运用Hough变换检测出车牌水平和垂直倾斜度后，再进一步进行字符分割，具有较好的效果。     

基于快速广义Hough变换的倾斜车牌定位

提出一种倾斜车牌定位算法。算法使用快速广义Hough变换检测车牌倾斜角度,并检测出倾斜车牌在图像平面中的长宽比例和车牌位置,定位出倾斜车牌。仿真实验表明该算法能正确定位倾斜车牌。     

基于Gauss消元法的车牌图像变形矫正

车辆牌照识别是智能交通系统的重要组成部分，而车牌图像的分割定位与字符提取是车牌识别系统的关键步骤，定位提取效果直接决定了车牌识别系统的工作效率。斜向拍摄的车牌图像存在着透视变形，该情况下直接进行车牌旋转不能有效地矫正，必须进行变形矫正才能准确地提取出各个牌照字符。文中利用双线性空间映射来矫正变形车牌图像，采用高斯消元法来计算该映射方程组的解，从而较好地解决了车牌透视变形的矫正问题，提高了车牌图像变形矫正的运算精度与速度。     

基于垂直线条密度质心法的车牌倾斜校正

车牌倾斜校正是车牌自动识别系统中的重要技术之一。提出了基于垂直线条密度质心法的汽车牌照倾斜校正的新方法。首先分析几种常见的车牌倾斜模式,然后先除去车牌边框及一些污点,最后把车牌图像划分成N列垂直线条,求出每列像素的质心,利用最小二乘法把每列质心拟合成一条直线,并推导出其斜率,由此确定出车牌的倾斜角度。文中给出了实验结果及分析,并与传统的Hough变换法和旋转投影法相比较,结果表明该方法精确、抗干扰性强、快速有效。     

基于商空间理论的车牌定位

1引言 随着智能交通系统的全面实施,汽车车牌的自动识别技术变得越来越重要,而牌照定位又是车牌识别中的关键环节.由于受天气、背景、磨损等外界干扰因素的影响,造成得到的车牌图像模糊,牌照区域不明显,给牌照区域的提取带来了较大的困难,因此牌照定位一直是汽车牌照识别系统中的重点和难点.在使用广泛的基于灰度图像的识别中,常用的方法有基于牌照纹理特征对图像进行检测,使用遗传算法对图像进行优化搜索,使用Hough变换以检测直线来提取车牌边界区域等.这些方法都有各自的优点,但是在实际应用中也存在一些不足之处,因此仍然需要做进一步的改进.     

Real-time detection of lines using parallel coordinates and CUDA

The Hough transform is a well-known and popular algorithm for detecting lines in raster images. The standard Hough transform is rather slow to be usable in real time, so different accelerated and approximated algorithms exist. This study proposes a modified accumulation scheme for the Hough transform, using a new parameterization of lines “PClines”. This algorithm is suitable for computer systems with a small but fast read-write memory, such as today’s graphics processors. The algorithm requires no floating-point computations or goniometric functions. This makes it suitable for special and low-power processors and special-purpose chips. The proposed algorithm is evaluated both on synthetic binary images and on complex real-world photos of high resolutions. The results show that using today’s commodity graphics chips, the Hough transform can be computed at interactive frame rates, even with a high resolution of the Hough space and with the Hough transform fully computed.     

Multiresolution Hough transform-an efficient method of detectingpatterns in images

A new multiresolution coarse-to-fine search algorithm for efficient computation of the Hough transform is proposed. The algorithm uses multiresolution images and parameter arrays. Logarithmic range reduction is proposed to achieve faster convergence. Discretization errors are taken into consideration when accumulating the parameter array. This permits the use of a very simple peak detection algorithm. Comparative results using three peak detection methods are presented. Tests on synthetic and real-world images show that the parameters converge rapidly toward the true value. The errors in ρ and &thetas;, as well as the computation time, are much lower than those obtained by other methods. Since the multiresolution Hough transform (MHT) uses a simple peak detection algorithm, the computation time will be significantly lower than other algorithms if the time for peak detection is also taken into account. The algorithm can be generalized for patterns with any number of parameters     

A convolution approach to the circle Hough transform for arbitrary radius

The Hough transform is a well-established family of algorithms for locating and describing geometric figures in an image. However, the computational complexity of the algorithm used to calculate the transform is high when used to target complex objects. As a result, the use of the Hough transform to find objects more complex than lines is uncommon in real-time applications. We describe a convolution method for calculating the Hough transform for finding circles of arbitrary radius. The algorithm operates by performing a three-dimensional convolution of the input image with an appropriate Hough kernel. The use of the fast Fourier transform to calculate the convolution results in a Hough transform algorithm with reduced computational complexity and thus increased speed. Edge detection and other convolution-based image processing operations can be incorporated as part of the transform, which removes the need to perform them with a separate pre-processing or post-processing step. As the Discrete Fourier Transform implements circular convolution rather than linear convolution, consideration must be given to padding the input image before forming the Hough transform.     

复杂背景下的车牌定位和字符分割研究

提出了一种综合边缘检测、投影特征的车牌定位方法和基于垂直投影及模板匹配的字符分割方法,提取车牌灰度图像边缘,实验结果显示该算法检测边缘的速度快,车牌区域轮廓清晰,采用投影法确定车牌区域,用HOUGH变换检测倾斜角度进而对倾斜的车牌进行矫正,通过字符分割算法对车牌字符进行切割,有效地解决了复杂环境的干扰、车牌尺寸变化等问题。对不同背景下的光照车牌进行了大量实验,结果表明该算法能准确地进行车牌定位以及字符分割,具有较好的鲁棒性。     

An improved constant-time algorithm for computing the Radon andHough transforms on a reconfigurable mesh

The Hough transform is an important problem in image processing and computer vision. An efficient algorithm for computing the Hough transform has been proposed on a reconfigurable array by Kao et al. (1995). For a problem with an √N×√N image and an n×n parameter space, the algorithm runs in a constant time on a three-dimensional (3-D) n×n×N reconfigurable mesh where the data bus is N¹c/-bit wide. To our best knowledge, this is the most efficient constant-time algorithm for computing the Hough transform on a reconfigurable mesh. In this paper, an improved Hough transform algorithm on a reconfigurable mesh is proposed. For the same problem, our algorithm runs in constant time on a 3-D n*n×n×√n√n reconfigurable mesh, where the data bus is only log N-bit wide. In most practical situations, n=O(√N). Hence, our algorithm requires much less VLSI area to accomplish the same task. In addition, our algorithm can compute the Radon transform (a generalized Hough transform) in O(1) time on the same model, whereas the algorithm in the above paper cannot be adapted to computing Radon transform easily     

基于Hough变换的图象检测对偶点法

提出了图形边界对偶点的概念,证明了对对称图形对偶点的存在性及其关于平移、缩放和旋转组合变换的不变性;结合Hough变换,建立了一种快速的图象检测方法,详细给出圆和椭圆检测的对偶点算法,并补充了文献[1]中公式的不足,实验结果表明,该对偶点法除保留了Hough变换法的容错性和鲁棒性外,还具有很快的计算速度,而且可以灵活地应用于解决一些较难的图象检测问题。     

Accurate and efficient curve detection in images: the importance sampling Hough transform

The Hough transform is a well known technique for detecting parametric curves in images. We place a particular group of Hough transforms, the probabilistic Hough transforms, in the framework of importance sampling. This framework suggests a way in which probabilistic Hough transforms can be improved: by specifying a target distribution and weighting the sampled parameters accordingly to make identification of curves easier. We investigate the use of clustering techniques to simultaneously identify multiple curves in the image. We also use probabilistic arguments to develop stopping conditions for the algorithm. Results from applying our method and two popular versions of the Hough transform to both simulated and real data are shown.     

Image processing on the OTIS-mesh optoelectronic computer

We develop algorithms for histogramming, histogram modification, Hough transform, and image shrinking and expanding on an OTIS-mesh optoelectronic computer. Our algorithm for the Hough transform is based upon a mesh algorithm for the Hough transform which is also developed in this paper. This new mesh algorithm improves upon the previous mesh Hough transform algorithms     

On the inverse Hough transform

In this paper, an inverse Hough transform algorithm is proposed. This algorithm reconstructs correctly the original image, using only the data of the Hough transform space and it is applicable to any binary image. As a first application, the inverse Hough transform algorithm is used for straight-line detection and filtering. The lines are detected not just as continuous straight lines, which is the case of the standard Hough transform, but as they really appear in the original image, i.e., pixel by pixel. To avoid the quantization effects in the Hough transform space, inversion conditions are defined, which are associated only with the dimensions of the images. Experimental results indicate that the inverse Hough transform algorithm is robust and accurate     

Genetic Algorithm for the Extraction of Nonanalytic Objects from Multiple Dimensional Parameter Space

A new approach of the Hough transform is proposed, which makes use of the genetic searching algorithm. By using this proposed algorithm, we can resolve the main obstacle of the Hough transform, which demands an enormous amount of storage for the Hough space. The idea of this genetic Hough technique is applicable to the recognition of both analytic and nonanalytic patterns. Based on the analysis of peak formation in the 4D generalized Hough transform's parameter space, a fitness function is derived, which represents the statistical weight of the existence of desired objects. By using the genetic approach to extract peaks in the parameter space, the physical storage for the 4D Hough parameter domain is not required during the detection while the accuracy of the detected parameters can be significantly improved.