基于ASIFT道路交通标志的定位与识别方法
|
摘要
针对智能交通系统中的交通标志识别,提出了一种基于ASIFT(仿射尺度不变特征变换)算法的定位与识别方法。首先,分类道路交通标志并提取模板;然后,采用ASIFT算法选择目标图像对应的模板和交通标志集合;其次,根据匹配模板ASIFT特征点的几何分布,定位交通标志的目标区域;最后,计算目标区域和交通标志集合的特征向量欧氏距离。基于ASIFT算法,达到了准确的定位与识别效果。通过实验,验证了该方法的可行性。
Aiming at traffic sign recognition in ITS( Intelligent Transport System),an ASIFT-based( Affine-ScaleInvariant Feature Transform) location and recognition method is proposed. Firstly,road traffic signs are classified and templates are extracted; secondly,the template and traffic signs set,matched with target image,are selected based on ASIFT; then,according to geometric distribution of ASIFT feature points in the matching template,target area is located on the target image; finally,Euclidean distance of feature vector is calculated between target area and one of images in traffic signs set. Based on ASIFT algorithm,accurate location and recognition effect come true. The feasibility of the method is verified by experiments.
Aiming at traffic sign recognition in ITS( Intelligent Transport System),an ASIFT-based( Affine-ScaleInvariant Feature Transform) location and recognition method is proposed. Firstly,road traffic signs are classified and templates are extracted; secondly,the template and traffic signs set,matched with target image,are selected based on ASIFT; then,according to geometric distribution of ASIFT feature points in the matching template,target area is located on the target image; finally,Euclidean distance of feature vector is calculated between target area and one of images in traffic signs set. Based on ASIFT algorithm,accurate location and recognition effect come true. The feasibility of the method is verified by experiments.
引文
[1]李祥熙,张航.交通标志识别研究综述[J].公路交通科技应用技术版,2010(6):253-257.
[2]Morel J M,Yu G.ASIFT:A New Framework for Fully Affine Invariant Image Comparison[J].SIAM Journal on Imaging Sciences,2009,2(2):438-469.
[3]Lowe D G.Distinctive Image Features from Scale-Invariant Keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
[4]郑永斌,黄新生,丰松江.SIFT和旋转不变LBP相结合的图像匹配算法[J].计算机辅助设计与图形学学报,2010,2(2):286-292.
[5]蒋庆斌,王浩,赵力.基于UKF与SIFT的车辆跟踪算法研究[J].计算机测量与控制,2012,20(10):2824-2826.
[6]王鹏宇.基于小波和色彩传递夜视图像彩色融合算法研究[J].电子器件,2014,37(1):143-145.
[7]刘欢,郝矿荣,丁永生,等.光照鲁棒的抗模糊新组合不变矩图像匹配方法[J].传感技术学报,2013,26(9):1258-1264.
[8]傅卫平,秦川,刘佳,等.基于SIFT算法的图像目标匹配与定位[J].仪器仪表学报,2011,32(1):163-169.
[9]温文雅.一种改进的Harris多尺度角点检测算法[J].重庆理工大学学报,2012,26(8):94-97.
[10]陈杏环,闫海煜.基于随机游走策略改进的降雪模型图像分割方法[J].电子器件,2014,37(2):350-354.
[11]高绍雷,沈建新,周洪亚,等.高斯核函数在角膜地形图仪图像处理中的应用[J].小型微型计算机系统,2011,32(10):2074-2077.
[12]田莹,徐超,张一鸣.基于Gabor小波和ASIFT特征点的人耳识别[J].计算机应用与软件,2013,30(11):202-205.
[13]张良,王海丽,吴仁彪.基于改进局部不变特征的兴趣点匹配[J].电子与信息学报,2009,31(11):2620-2625.
[14]周强,杨智勇,孙小燕,等.基于图像欧式距离的KPCA SAR图像目标识别算法[J].火力与指挥控制,2012,37(12):38-41.
[15]张闯,王婷婷,孙东娇,等.基于欧式距离图的图像边缘检测[J].中国图象图形学报,2013,18(2):176-183.
[16]唐朝伟,肖健,邵艳清,等.全局结构化SIFT描述子在图像匹配中的应用[J].华中科技大学学报:自然科学版,2012,40(1):15-20.
[2]Morel J M,Yu G.ASIFT:A New Framework for Fully Affine Invariant Image Comparison[J].SIAM Journal on Imaging Sciences,2009,2(2):438-469.
[3]Lowe D G.Distinctive Image Features from Scale-Invariant Keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
[4]郑永斌,黄新生,丰松江.SIFT和旋转不变LBP相结合的图像匹配算法[J].计算机辅助设计与图形学学报,2010,2(2):286-292.
[5]蒋庆斌,王浩,赵力.基于UKF与SIFT的车辆跟踪算法研究[J].计算机测量与控制,2012,20(10):2824-2826.
[6]王鹏宇.基于小波和色彩传递夜视图像彩色融合算法研究[J].电子器件,2014,37(1):143-145.
[7]刘欢,郝矿荣,丁永生,等.光照鲁棒的抗模糊新组合不变矩图像匹配方法[J].传感技术学报,2013,26(9):1258-1264.
[8]傅卫平,秦川,刘佳,等.基于SIFT算法的图像目标匹配与定位[J].仪器仪表学报,2011,32(1):163-169.
[9]温文雅.一种改进的Harris多尺度角点检测算法[J].重庆理工大学学报,2012,26(8):94-97.
[10]陈杏环,闫海煜.基于随机游走策略改进的降雪模型图像分割方法[J].电子器件,2014,37(2):350-354.
[11]高绍雷,沈建新,周洪亚,等.高斯核函数在角膜地形图仪图像处理中的应用[J].小型微型计算机系统,2011,32(10):2074-2077.
[12]田莹,徐超,张一鸣.基于Gabor小波和ASIFT特征点的人耳识别[J].计算机应用与软件,2013,30(11):202-205.
[13]张良,王海丽,吴仁彪.基于改进局部不变特征的兴趣点匹配[J].电子与信息学报,2009,31(11):2620-2625.
[14]周强,杨智勇,孙小燕,等.基于图像欧式距离的KPCA SAR图像目标识别算法[J].火力与指挥控制,2012,37(12):38-41.
[15]张闯,王婷婷,孙东娇,等.基于欧式距离图的图像边缘检测[J].中国图象图形学报,2013,18(2):176-183.
[16]唐朝伟,肖健,邵艳清,等.全局结构化SIFT描述子在图像匹配中的应用[J].华中科技大学学报:自然科学版,2012,40(1):15-20.