广西师范大学学报（自然科学版） ›› 2023, Vol. 41 ›› Issue (3): 41-52.doi: 10.16088/j.issn.1001-6600.2022060202

• 研究论文 • 上一篇    下一篇

一种面向无人机航拍图像的快速拼接算法

梁镇锋, 夏海英*   

  1. 广西师范大学 电子工程学院, 广西 桂林 541004
  • 收稿日期:2022-06-02 修回日期:2022-08-30 出版日期:2023-05-25 发布日期:2023-06-01
  • 通讯作者: 夏海英(1983—), 女, 山东聊城人, 广西师范大学教授, 博士。E-mail: xhy22@mailbox.gxnu.edu.cn
  • 基金资助:
    国家自然科学基金(61762014, 62106054); 广西科技重大专项(桂科AA20302003)

A Fast Stitching Algorithm for UAV Aerial Images

LIANG Zhenfeng, XIA Haiying*   

  1. College of Electronic Engineering, Guangxi Normal University, Guilin Guangxi, 541004, China
  • Received:2022-06-02 Revised:2022-08-30 Online:2023-05-25 Published:2023-06-01

PDF (PC)

50

摘要： 针对现有图像拼接方法对分辨率高、影像信息复杂的无人机航拍图像处理速度慢、匹配精度差的问题,提出一种面向无人机航拍图像的快速拼接算法。首先,利用FM(Fourier-Mellin)算法求出图像的重叠区域,并获取图像重叠区域内的感兴趣区域,结合区域分块以及图像信息熵得到最终的特征提取区域,设置为图像掩膜;接着,在特征提取区域提取特征点,并且对特征点匹配对进行两轮筛选,减少RANSAC算法迭代次数;最后用改进的RANSAC算法进行误匹配点对的剔除以筛选出准确度较高的匹配点对。比较该算法与其他算法的运行效率以及图像的拼接质量等相关指标。实验结果显示该算法较SURF、SIFT、区域分块算法,航拍图像拼接时间分别降低35%、56%、57%,满足航拍图像对拼接精度以及实时性的要求。

关键词: 航拍图像, 图像拼接, 特征提取区域, 傅里叶-梅林变换, 图像掩膜

Abstract: Aiming at the problems of slow processing speed and poor matching accuracy of the existing image stitching methods for UAV aerial images with high resolution and complex image information, a fast stitching algorithm for UAV aerial images is proposed. Firstly, the FM (Fourier-Mellin) algorithm is used to find out the overlapping region of the image, and obtain ROI from the overlap area of image, combining the area blocking and the image entropy to get the region of final feature extraction as the image mask. Then, the feature points are extracted in the area of feature extraction and two rounds of filtering are performed on the feature matching pairs to reduce the number of RANSAC algorithm iterations. Finally, an improved RANSAC algorithm is used to reject the wrong pairs in order to filter out the pairs with higher accuracy. Compared the operation efficiency of this algorithm with other algorithms, as well as the image stitching quality and other related indexes, the experimental results show that this algorithm reduces the stitching time by 35%, 56% and 57% compared with SURF, SIFT and area blocking algorithms respectively, meeting the requirements of today's aerial images for stitching accuracy and real-time performance.

Key words: aerial images, image stitching, feature extraction area, Fourier-Mellin transform, image mask

中图分类号:  V19;TP391.41

[1] 陈林奇, 李廷会. 基于双空间PSO算法的四旋翼无人机自抗扰控制器优化设计[J]. 广西师范大学学报(自然科学版), 2019, 37(3):42-49. DOI: 10.16088/j.issn.1001-6600.2019.03.005.
[2] 雷荣智,杨维芳,苏小宁.基于改进YOLO-v3的无人机遥感图像农村地物分类[J].电子设计工程,2023,31(3):178-184. DOI: 10.14022/j.issn1674-6236.2023.03.037.
[3] 洪梓铭.电力巡检的移动作业无人机倾斜影像三维采集模型[J].电子设计工程,2023,31(6):149-152,157. DOI: 10.14022/j.issn1674-6236.2023.06.031.
[4] 孟现彪,刘盛庆,史雅茹,等.无人机激光雷达高程测量精度评定分析[J].桂林理工大学学报,2022,42(2):438-442.
[5] 徐俊辉,欧阳缮,廖可非,等.基于相关检测的旋翼无人机检测与识别方法[J].桂林电子科技大学学报,2021,41(6):431-436. DOI: 10.16725/j.cnki.cn45-1351/tn.2021.06.002.
[6] 李加亮, 蒋品群, 夏海英. 基于网格变形和余弦函数权重的图像拼接方法[J]. 广西师范大学学报(自然科学版), 2020, 38(4):42-53. DOI: 10.16088/j.issn.1001-6600.2020.04.005.
[7] BROWN M, LOWE D G. Automatic panoramic image stitching using invariant features[J]. International Journal of Computer Vision, 2007, 74(1): 59-73. DOI: 10.1007/s11263-006-0002-3.
[8] RUBLEE E, RABAUD V, KONOLIGE K, et al. ORB: an efficient alternative to SIFT or SURF[C]//2011 International Conference on Computer Vision. Piscataway, NJ: IEEE, 2011: 2564-2571. DOI: 10.1109/ICCV.2011.6126544.
[9] BAY H, ESS A, TUYTELAARS T, et al. Speeded-up robust features (SURF)[J]. Computer Vision and Image Understanding, 2008, 110(3): 346-359. DOI: 10.1016/j.cviu.2007.09.014.
[10] 张红民, 张见双, 罗永涛, 等. 一种基于图像区域分块的 SIFT 快速配准方法[J]. 红外技术, 2017, 39(4): 341-344.
[11] ZARAGOZA J, CHIN T J, TRAN Q H, et al. As-projective-as-possible image stitching with moving DLT[J]. IEEE Transaction on Pattern Analysis and Machine Intelligence, 2014, 36(7):1285-1298. DOI: 10.1109/TPAMI.2013.247.
[12] LIN C C, PANKANTI S U, RAMAMURTHY K N, et al. Adaptive as-natural-as-possible image stitching[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Piscataway, NJ: IEEE, 2015: 1155-1163. DOI: 10.1109/CVPR.2015.7298719.
[13] CHANG C H, SATO Y, CHUANG Y Y. Shape-preserving half-projective warps for image stitching[C]//2014 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2014: 3254-3261. DOI: 10.1109/CVPR.2014.422.
[14] CHEN Y S, CHUANG Y Y. Natural image stitching with the global similarity prior[C]//Computer Vision-ECCV 2016. Cham: Springer, 2016: 186-201. DOI: 10.1007/978-3-319-46454-1_12.
[15] 杨明东, 石英, 华逸伦, 等. 基于匹配策略融合的低误差快速图像拼接算法[J]. 计算机应用研究, 2019, 36(4): 1222-1227. DOI: 10.19734/j.issn.1001-3695.2017.11.0772.
[16] LIN K M, JIANG N J, CHEONG L F, et al. SEAGULL: seam-guided local alignment for parallax-tolerant image stitching[C]//Computer Vision-ECCV 2016. Cham: Springer, 2016: 370-385. DOI: 10.1007/978-3-319-46487-9_23.
[17] JIA Q, LI Z J, FAN X, et al. Leveraging line-point consistence to preserve structures for wide parallax image stitching[C]//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Piscataway, NJ: IEEE, 2021: 12181-12190. DOI: 10.1109/CVPR46437.2021.01201.
[18] NIE L, LIN C Y, LIAO K, et al. A view-free image stitching network based on global homography[J]. Journal of Visual Communication and Image Representation, 2020, 73: 102950. DOI: 10.1016/j.jvcir.2020.102950.
[19] NIE L, LIN C Y, LIAO K, et al. Unsupervised deep image stitching: reconstructing stitched features to images[J]. IEEE Transactions on Image Processing, 2021, 30: 6184-6197. DOI: 10.1109/TIP.2021.3092828.
[20] REDDY B S, CHATTERJI B N. An FFT-based technique for translation, rotation, and scale-invariant image registration[J]. IEEE Transactions on Image Processing, 1996, 5(8): 1266-1271. DOI: 10.1109/83.506761.
[21] 周晓炜, 赵琰. 基于变换域和形状特征的感知图像哈希算法[J]. 计算机应用与软件, 2021, 38(5):218-224. DOI: 10.3969/j.issn.1000-386x.2021.05.035.
[22] B ZITOVÁ, FLUSSER J. Image registration methods: a survey[J]. Image and Vision Computing, 2003, 21(11):977-1000. DOI: 10.1016/S0262-8856(03)00137-9.
[23] 刘洋, 王利民, 孙铭会. 基于信息熵函数的启发式贝叶斯因果推理[J]. 计算机学报, 2021, 44(10):2135-2147. DOI: 10.11897/SP.J.1016.2021.02135.
[24] SINGHA ROY S, BARMAN S. A non-invasive cancer gene detection technique using FLANN based adaptive filter[J]. Microsystem Technologies, 2021, 27(2): 463-478. DOI: 10.1007/s00542-018-4036-6.
[25] 刘淑伟, 陈威, 赵伟,等. 基于簇内乘积量化的最近邻检索方法[J]. 计算机学报, 2020, 43(2):303-314. DOI: 10.11897/SP.J.1016.2020.00303.
[26] 吴禄慎, 陈小杜. 一种改进AKAZE特征和RANSAC的图像拼接算法[J]. 计算机工程, 2021, 47(1):246-254. DOI: 10.19678/j.issn.1000-3428.0056874.
[27] SUN J M, SHEN Z H, WANG Y, et al. LoFTR: detector-free local feature matching with transformers[C]//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Piscataway, NJ: IEEE, 2021: 8918-8927. DOI: 10.1109/CVPR46437.2021.00881.
[28] 呼亚萍, 孔韦韦, 李萌,等. 改进TV图像去噪模型的全景图像拼接算法[J]. 计算机工程与应用, 2021,57(17):203-209. DOI: 10.3778/j.issn.1002-8331.2005-0216.
[1] 李加亮, 蒋品群, 夏海英. 基于网格变形和余弦函数权重的图像拼接方法[J]. 广西师范大学学报（自然科学版）, 2020, 38(4): 42-53.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
[1] 程瑞, 何明先, 钟春英, 罗树毅, 武正军. 野生与人工繁育鳄蜥游泳能力比较[J]. 广西师范大学学报（自然科学版）, 2021, 39(1): 79 -86 .
[2] 陈丹妮, 陈志林, 周善义. 中国蚁科昆虫名录——切叶蚁亚科(补遗)[J]. 广西师范大学学报（自然科学版）, 2021, 39(1): 87 -97 .
[3] 徐国良. 广东7种蕨类植物新记录[J]. 广西师范大学学报（自然科学版）, 2021, 39(1): 114 -118 .
[4] 何韩吉, 邓光明, 葛梦兰. 中原城市群空气质量空间关联研究[J]. 广西师范大学学报（自然科学版）, 2021, 39(3): 151 -162 .
[5] 谢冬林, 邓国和. 随机利率跳扩散模型下幂型乘积远期生效期权定价[J]. 广西师范大学学报（自然科学版）, 2021, 39(5): 158 -172 .
[6] 蒋向辉, 谭荣, 杨永平, 肖清淙. 十大功劳甘草汤治疗肝炎的网络药理学研究[J]. 广西师范大学学报（自然科学版）, 2021, 39(5): 198 -209 .
[7] 许伦辉, 林世城. 基于分治思想的扫地机器人全覆盖路径规划算法研究[J]. 广西师范大学学报（自然科学版）, 2021, 39(6): 54 -62 .
[8] 周子豪, 刘禹含, 谭艳红, 蒙玉清, 巫虹颖, 黄锦龙, 武正军. 酶法制备福寿螺内脏团抗菌肽[J]. 广西师范大学学报（自然科学版）, 2021, 39(6): 154 -161 .
[9] 闫莎. 一类三种群捕食者-食饵交错扩散模型的整体解[J]. 广西师范大学学报（自然科学版）, 2021, 39(2): 125 -131 .
[10] 彭涛, 唐经, 何凯, 胡新荣, 刘军平, 何儒汉. 基于多步态特征融合的情感识别[J]. 广西师范大学学报（自然科学版）, 2022, 40(3): 104 -111 .
版权所有 © 广西师范大学学报（自然科学版）编辑部
地址：广西桂林市三里店育才路15号　邮编：541004
电话：0773-5857325　E-mail: gxsdzkb@mailbox.gxnu.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发