反馈机制下图像镶嵌累积误差处理方法的研究
|
摘要
针对多幅图像镶嵌中累积误差的问题,提出一种基于反馈机制的减小累积误差的图像镶嵌方法.在图像镶嵌之前,将所有待镶嵌图像进行柱面投影.算法对待拼接图像进行尺度不变特征点提取,在粗匹配的基础上,对互匹配唯一性约束原则和随机抽样一致方法建立一种反馈机制.运用最佳重叠度原理尽可能找出更多正确的匹配点对,为图像变换矩阵的计算提供良好的数据基础.实验结果表明:该种机制能够从本质上提高图像变换矩阵的精度,从而有效地减少多幅图像镶嵌的累积误差,并且对不同图像序列有着良好的自适应性.
To solve the problem of accumulated error in multiple image mosaic,this paper proposed a new image mosaic method based on Feedback mechanism to reduce accumulated error. All images will be cylindrical projected before mosaic. The algorithm takes images and find the SIFT feature points. On the basis of rough match,this paper established a feedback mechanism on the principle of only constraint mutual match. Rely on this kind of mechanism and best overlapping principle,the proposed algorithm could find out more correctly matching points,which provide a good basis for the calculation of the transformation matrix. The result show s that the feedback mechanism can effectively increase the accuracy of the image transformation matrix,which can reduce the accumulated error of multiple images mosaic. Additionally,this mechanism has good adaptability for different set of images.
To solve the problem of accumulated error in multiple image mosaic,this paper proposed a new image mosaic method based on Feedback mechanism to reduce accumulated error. All images will be cylindrical projected before mosaic. The algorithm takes images and find the SIFT feature points. On the basis of rough match,this paper established a feedback mechanism on the principle of only constraint mutual match. Rely on this kind of mechanism and best overlapping principle,the proposed algorithm could find out more correctly matching points,which provide a good basis for the calculation of the transformation matrix. The result show s that the feedback mechanism can effectively increase the accuracy of the image transformation matrix,which can reduce the accumulated error of multiple images mosaic. Additionally,this mechanism has good adaptability for different set of images.
引文
[1]Ye Si,Li Liang-fu,Jiao Ting.A ghost removing method for panoramic image mosaic w ith moving objects[J].Journal of Shanxi Normal University(Natural Science Edition),2014,4(12):5-11.
[2]Gong A-du,He Xiao-ying,Lei Tian-jie.Fast image processing method of UAV w ithout control data[J].Journal of Geo-Information Science,2010,12(2):254-260.
[3]Peng Xing-hong,Song Hong-zhi,Zou Xian-jun.Improved panoramic image mosaic algorithm[J].Computer Engineering and Applications,2010,46(25):195-198.
[4]Li Cong-li,Han Yu-sheng,Leng Xiao-yan.Elimination of pseudo matching in image stitching[J].Computer Applications and Softw are,2006,23(11):93-95.
[5]He Jing,Li Yong-shu,Lu Heng.Experimental study on UAV image stitching error[J].Laser&Optoelectronics Progress,2012,48(12):60-64.
[6]David G L.Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision 2004,60(2):91-110.
[7]Li Zeng,Zhang Si,Zhang Jie,et al.Dynamic image mosaic via SIFT and dynamic programming[J].M achine Vision and Applications,2014,25(5):1271-1282.
[8]Li Jin,Wang Yan-wei,Li Hong.Image Mosaic based on simplified SIFT[C].International Conference on M echanical Engineering and Automation Advances in Biomedical Engineering,2012:100-106.
[9]Li Yan.Digital image mosaic technology based on improved genetic algorithm[J].Journal of M ultimedia,2014,9(3):428-434.
[10]Zhu Jun,Ren Ming-wu,Liu Qing-shan.Image mosaic method based on SIFT features of line segment[J].Computational and M athematical M ethods in M edicine,2014,14(8):11-16.
[11]Li Xiao-juan,Xi Xiao-yan,Zang Yi-hua,et al.Fusion algorithm based on particle sw arm optimization edge[J].Softw are Guide,2014,23(5):79-81.
[12]Xie Kai,Guo Heng-ye,Zhang Tian-wen.A survey of image mosaics technology[J].Acta Electronica Sinica,2004,32(4):630-634.
[13]Yu Hong-sheng,Jin Wei-qi.Evolvement of research on digital image mosaics methods[J].Infrared Technology,2009,31(6):348-353.
[14]Pérez J,Espinosa J,Vázquez C,et al.Retinal image quality assessment through a visual similarity index[J].Journal of M odern Optics,2013,60(7):544-550.
[1]叶偲,李良福,焦婷.全景图像镶嵌中运动目标的重影处理方法[J].陕西师范大学学报(自然科学版),2014,4(12):5-11.
[2]宫阿都,何孝莹,雷添杰,等.无控制点数据的无人机影像快速处理[J].地球信息科学学,2010,12(2):254-260.
[3]彭红星,宋鸿陟,邹湘军,等.改进的全景图拼接融合算法[J].计算机工程与应用,2010,46(25):195-198.
[4]李从利,韩裕生,冷晓艳,等.序列图像拼接过程中误匹配的消除[J].计算机应用与软件,2006,23(11):93-95.
[5]何敬,李永树,鲁恒.无人机影像拼接误差实验研究[J].激光与光电子学进展,2012,48(12):60-64.
[2]Gong A-du,He Xiao-ying,Lei Tian-jie.Fast image processing method of UAV w ithout control data[J].Journal of Geo-Information Science,2010,12(2):254-260.
[3]Peng Xing-hong,Song Hong-zhi,Zou Xian-jun.Improved panoramic image mosaic algorithm[J].Computer Engineering and Applications,2010,46(25):195-198.
[4]Li Cong-li,Han Yu-sheng,Leng Xiao-yan.Elimination of pseudo matching in image stitching[J].Computer Applications and Softw are,2006,23(11):93-95.
[5]He Jing,Li Yong-shu,Lu Heng.Experimental study on UAV image stitching error[J].Laser&Optoelectronics Progress,2012,48(12):60-64.
[6]David G L.Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision 2004,60(2):91-110.
[7]Li Zeng,Zhang Si,Zhang Jie,et al.Dynamic image mosaic via SIFT and dynamic programming[J].M achine Vision and Applications,2014,25(5):1271-1282.
[8]Li Jin,Wang Yan-wei,Li Hong.Image Mosaic based on simplified SIFT[C].International Conference on M echanical Engineering and Automation Advances in Biomedical Engineering,2012:100-106.
[9]Li Yan.Digital image mosaic technology based on improved genetic algorithm[J].Journal of M ultimedia,2014,9(3):428-434.
[10]Zhu Jun,Ren Ming-wu,Liu Qing-shan.Image mosaic method based on SIFT features of line segment[J].Computational and M athematical M ethods in M edicine,2014,14(8):11-16.
[11]Li Xiao-juan,Xi Xiao-yan,Zang Yi-hua,et al.Fusion algorithm based on particle sw arm optimization edge[J].Softw are Guide,2014,23(5):79-81.
[12]Xie Kai,Guo Heng-ye,Zhang Tian-wen.A survey of image mosaics technology[J].Acta Electronica Sinica,2004,32(4):630-634.
[13]Yu Hong-sheng,Jin Wei-qi.Evolvement of research on digital image mosaics methods[J].Infrared Technology,2009,31(6):348-353.
[14]Pérez J,Espinosa J,Vázquez C,et al.Retinal image quality assessment through a visual similarity index[J].Journal of M odern Optics,2013,60(7):544-550.
[1]叶偲,李良福,焦婷.全景图像镶嵌中运动目标的重影处理方法[J].陕西师范大学学报(自然科学版),2014,4(12):5-11.
[2]宫阿都,何孝莹,雷添杰,等.无控制点数据的无人机影像快速处理[J].地球信息科学学,2010,12(2):254-260.
[3]彭红星,宋鸿陟,邹湘军,等.改进的全景图拼接融合算法[J].计算机工程与应用,2010,46(25):195-198.
[4]李从利,韩裕生,冷晓艳,等.序列图像拼接过程中误匹配的消除[J].计算机应用与软件,2006,23(11):93-95.
[5]何敬,李永树,鲁恒.无人机影像拼接误差实验研究[J].激光与光电子学进展,2012,48(12):60-64.