基于灰度方差及边缘密度的车牌定位研究.docVIP

文章题目 基于灰度方差和边缘密度的车牌定位算法 创 新 点 自 述 在车牌定位的过程中，由于光照变化、视点和距离变化、车辆运动、复杂背景等原因，摄像机很难获取到高质量的图像。为了克服这些问题对以后的车牌定位算法的影响，提出了利用车牌区域的灰度方差和边缘密度的性质来进行车牌定位的新算法。车牌区域具有灰度方差近似相等和边缘密度近似相等且在一定范围内的性质，利用这些性质来增强车牌区域。为了定位车牌区域，提出基于车牌区域边缘密度性质的匹配滤波器，该滤波器可以有效地提取出所有候选目标。利用形态学和先验知识进行目标真实性验证。在形态学运算中，采用自适应的方法设置机构元素，可以有效地提高算法的灵活性。通过对不同环境条件下获取的700幅图像进行实验，实验结果表明该算法有效地提高车牌区域的图像质量且车牌目标定位准确率达到98.4%，验证了算法的有效性和鲁棒性。在车牌定位的过程中，由于光照变化、视点和距离变化、车辆运动、复杂背景等原因，摄像机很难获取到高质量的图像。为了克服这些问题对以后的车牌定位算法的影响，提出了利用车牌区域的灰度方差和边缘密度的性质来进行车牌定位的新算法。车牌区域具有灰度方差近似相等和边缘密度近似相等且在一定范围内的性质，利用这些性质来增强车牌区域。为了定位车牌区域，提出基于车牌区域边缘密度性质的匹配滤波器，该滤波器可以有效地提取出所有候选目标。利用形态学和先验知识进行目标真实性验证。在形态学运算中，采用自适应的方法设置机构元素，可以有效地提高算法的灵活性。通过对不同环境条件下获取的700幅图像进行实验，实验结果表明该算法有效地提高车牌区域的图像质量且车牌目标定位准确率达到98.4%，验证了算法的有效性和鲁棒性。TP391.41 文章标识码：A Car license Plate location algorithm based on Intensity Variance and Edge Density Abstract：Inhe camera can’t obtain the high quality images easily due to illumination conditions, viewpoint and distance changes, vehicle motion, complex background, etc. In order to alleviate these problems to impact future car license plate location algorithm, a new algorithm using intensity variance and edge density in license plate region is proposed. The license plate region had the properties that the intensity variance is approximately equal and edge density also is approximately equal and within a certain range, then the license plate region is enhanced using these properties. In order to locate the license plate region, a matched filter based the property of edge density is designed. The filter can effectively extract all possible objects. Then, the object authenticity confirmation by morphological operation and priori knowledge is used. In the morphological operation, an adaptive method is used to set the structuring element; it can effectively improve the flexibility of the proposed algorithm. An experiment is undertaken with 700 actual images from different environments, and the experimental results show that the proposed algorithm can e