基于改进ORB的警用无人机图像配准方法
|
摘要
针对无人机航拍图像尺寸大、分辨率高、信息丰富等特点,提出一种基于改进ORB (oriented FAST and rotated BRIEF)算法的图像配准方法。在待配准图像中构造一个掩模,逐步移动掩模检测ORB特征点;采用非极大值抑制方法剔除聚集的特征点,对保留下的特征点进行ORB特征描述;使用汉明距离进行特征点匹配;采用PROSAC算法进行匹配对提纯并计算变换矩阵。实验结果表明,该方法对于存在旋转和尺度、模糊、视角、光照等变化情况的图像,在配准精度方面比ORB提高了3%-11%,在无人机航拍图像中具有较好的适应性。
For the UAV aerial images with large size,high resolution and rich information,an image registration method based on improved ORB(oriented FAST and rotated BRIEF)algorithm was proposed.A mask was constructed in the image to be registered,and moved gradually to detect feature points by ORB algorithm.The non-maximal suppression method was used to remove the clustered feature points,and the reserved feature points were described by ORB descriptor.The feature points were matched using Hamming distance.PROSAC algorithm was used to purify matches and calculate the transformation matrix.Experimental results show that for the images with changes of scale,rotation,blur,viewpoint and illumination,the proposed method can improve the registration accuracy by 3%-11% compared to ORB,and it has better applicability for UAV aerial images.
For the UAV aerial images with large size,high resolution and rich information,an image registration method based on improved ORB(oriented FAST and rotated BRIEF)algorithm was proposed.A mask was constructed in the image to be registered,and moved gradually to detect feature points by ORB algorithm.The non-maximal suppression method was used to remove the clustered feature points,and the reserved feature points were described by ORB descriptor.The feature points were matched using Hamming distance.PROSAC algorithm was used to purify matches and calculate the transformation matrix.Experimental results show that for the images with changes of scale,rotation,blur,viewpoint and illumination,the proposed method can improve the registration accuracy by 3%-11% compared to ORB,and it has better applicability for UAV aerial images.
引文
[1]REN Weijian, WANG Ziwei,KANG Chaohai.Remote sensing image of UAV registration based on improved SIFT algorithm[J].Computer Science,2015,42(s2):179-182(in Chinese).[任伟建,王子维,康朝海.基于改进SIFT算法的无人机遥感图像匹配[J].计算机科学,2015,42(s2):179-182.]
[2]LI Changchun,QI Xiudong,LEI Tianjie,et al.Research on unmanned aerial vehicle images mosaic quickly based on improved SURF algorithm[J].Geography and Geo-Information Science,2013,29(5):22-25(in Chinese).[李长春,齐修东,雷添杰,等.基于改进SURF算法的无人机遥感影像快速拼接[J].地理与地理信息科学,2013,29(5):22-25.]
[3]Adel E,Elmogy M,Elbakry H.Real time image mosaicing system based on feature extraction techniques[C]//International Conference on Computer Engineering&Systems.IEEE,2015:339-345.
[4]ZHANG Yunsheng,ZOU Zhengrong.Automatic registration method for remote sensing images based on improved ORB algorithm[J].Remote Sensing for Land&Resources,2013,25(3):20-24(in Chinese).[张云生,邹峥嵘.基于改进ORB算法的遥感图像自动配准方法[J].国土资源遥感,2013,25(3):20-24.]
[5]LIN Chunli,HAO Xuesen, MA Liancheng,et al.Partially occluded vehicle tracking based on ORB[J].Computer Engineering and Design,2016,37(1):242-246(in Chinese).[林春丽,郝学森,马连成,等.改进的ORB算法在有遮挡的车辆跟踪上的应用[J].计算机工程与设计,2016,37(1):242-246.]
[6]XU Xiaofan,WANG Yi,WANG Yongquan.Improved SIFT algorithm based on adaptive non-maximun suppression[J].Electronic Design Engineering,2014(18):180-182(in Chinese).[许晓帆,王毅,王永泉.基于自适应非极大值抑制的SIFT改进算法[J].电子设计工程,2014(18):180-182.]
[7]Rublee E,Rabaud V,Konolige K,et al.ORB:An efficient alternative to SIFT or SURF[C]//IEEE International Conference on Computer Vision,2011:2564-2571.
[8]Neubeck A,Gool LV.Efficient non-maximum suppression[C]//International Conference on Pattern Recognition.IEEE Computer Society,2006:850-855.
[9]DAI Xuemei,LANG Lang,CHEN Mengyuan.Research of image feature point matching based on improved ORB algorithm[J].Journal of Electronic Measurement and Instrument,2016,30(2):233-240(in Chinese).[戴雪梅,郎朗,陈孟元.基于改进ORB的图像特征点匹配研究[J].电子测量与仪器学报,2016,30(2):233-240.]
[10]XU Yang.Research on technologies of UAV remote sensing image mosaic[D].Nanjing:Nanjing University of Aeronautics and Astronautics,2012(in Chinese).[徐阳.无人机遥感图像拼接技术研究[D].南京:南京航空航天大学,2012.]
[11]ZHANG Chengtao.Research on algorithms for UAV aerial image stitching[D].Beijing:Beijing Institute of Technology,2015(in Chinese).[张成涛.无人机航拍图像拼接技术研究[D].北京:北京理工大学,2015.]
[12]LI Xiaohong,XIE Chengming,JIA Yizhen,et al.Rapid moving object detection algorithm based on ORB features[J].Journal of Electronic Measurement and Instrument,2013,27(5):455-460(in Chinese).[李小红,谢成明,贾易臻,等.基于ORB特征的快速目标检测算法[J].电子测量与仪器学报,2013,27(5):455-460.]
[2]LI Changchun,QI Xiudong,LEI Tianjie,et al.Research on unmanned aerial vehicle images mosaic quickly based on improved SURF algorithm[J].Geography and Geo-Information Science,2013,29(5):22-25(in Chinese).[李长春,齐修东,雷添杰,等.基于改进SURF算法的无人机遥感影像快速拼接[J].地理与地理信息科学,2013,29(5):22-25.]
[3]Adel E,Elmogy M,Elbakry H.Real time image mosaicing system based on feature extraction techniques[C]//International Conference on Computer Engineering&Systems.IEEE,2015:339-345.
[4]ZHANG Yunsheng,ZOU Zhengrong.Automatic registration method for remote sensing images based on improved ORB algorithm[J].Remote Sensing for Land&Resources,2013,25(3):20-24(in Chinese).[张云生,邹峥嵘.基于改进ORB算法的遥感图像自动配准方法[J].国土资源遥感,2013,25(3):20-24.]
[5]LIN Chunli,HAO Xuesen, MA Liancheng,et al.Partially occluded vehicle tracking based on ORB[J].Computer Engineering and Design,2016,37(1):242-246(in Chinese).[林春丽,郝学森,马连成,等.改进的ORB算法在有遮挡的车辆跟踪上的应用[J].计算机工程与设计,2016,37(1):242-246.]
[6]XU Xiaofan,WANG Yi,WANG Yongquan.Improved SIFT algorithm based on adaptive non-maximun suppression[J].Electronic Design Engineering,2014(18):180-182(in Chinese).[许晓帆,王毅,王永泉.基于自适应非极大值抑制的SIFT改进算法[J].电子设计工程,2014(18):180-182.]
[7]Rublee E,Rabaud V,Konolige K,et al.ORB:An efficient alternative to SIFT or SURF[C]//IEEE International Conference on Computer Vision,2011:2564-2571.
[8]Neubeck A,Gool LV.Efficient non-maximum suppression[C]//International Conference on Pattern Recognition.IEEE Computer Society,2006:850-855.
[9]DAI Xuemei,LANG Lang,CHEN Mengyuan.Research of image feature point matching based on improved ORB algorithm[J].Journal of Electronic Measurement and Instrument,2016,30(2):233-240(in Chinese).[戴雪梅,郎朗,陈孟元.基于改进ORB的图像特征点匹配研究[J].电子测量与仪器学报,2016,30(2):233-240.]
[10]XU Yang.Research on technologies of UAV remote sensing image mosaic[D].Nanjing:Nanjing University of Aeronautics and Astronautics,2012(in Chinese).[徐阳.无人机遥感图像拼接技术研究[D].南京:南京航空航天大学,2012.]
[11]ZHANG Chengtao.Research on algorithms for UAV aerial image stitching[D].Beijing:Beijing Institute of Technology,2015(in Chinese).[张成涛.无人机航拍图像拼接技术研究[D].北京:北京理工大学,2015.]
[12]LI Xiaohong,XIE Chengming,JIA Yizhen,et al.Rapid moving object detection algorithm based on ORB features[J].Journal of Electronic Measurement and Instrument,2013,27(5):455-460(in Chinese).[李小红,谢成明,贾易臻,等.基于ORB特征的快速目标检测算法[J].电子测量与仪器学报,2013,27(5):455-460.]