基于图像检索与GPS定位相结合的地标识别系统
|
摘要
为了改善地标识别系统识别准确度差与耗时较长的问题,提出了一种使用图像检索与GPS定位相结合的地标识别系统。讨论了尺度不变特征变换(SIFT)算法和加速稳健特征(SURF)算法的优缺点,并将这2种算法与改进的算法在检测的精确度和时间方面进行了比较研究。针对匹配过程中出现的误匹配问题,采用随机抽样一致性(RANSAC)算法剔除误匹配,达到了提纯的效果。实验结果表明:本文采用定位信息与SURF算法相结合的地标识别方法不仅提高了系统的检测正确率,而且还提高了系统的检索速度。
In this paper,a landmark recognition system based on image retrieval and GPS is proposed to address the problem that such systems being used at present tends to be inaccurate and time-consuming. The paper discussed the advantages and disadvantages of scale-invariant feature transform(SIFT) and speeded up robust features(SURF) algorithm. The two algorithms were compared with the improved algorithm in aspects of the detection accuracy and time. To cope with the mismatch in the matching process,random sample consensus(RANSAC) algorithm was used to eliminate the mismatch to achieve the purification effect. Results of the experiment show that the landmark recognition method which combines location information with SURF algorithm improves the detection accuracy and the retrieval speed of the system.
In this paper,a landmark recognition system based on image retrieval and GPS is proposed to address the problem that such systems being used at present tends to be inaccurate and time-consuming. The paper discussed the advantages and disadvantages of scale-invariant feature transform(SIFT) and speeded up robust features(SURF) algorithm. The two algorithms were compared with the improved algorithm in aspects of the detection accuracy and time. To cope with the mismatch in the matching process,random sample consensus(RANSAC) algorithm was used to eliminate the mismatch to achieve the purification effect. Results of the experiment show that the landmark recognition method which combines location information with SURF algorithm improves the detection accuracy and the retrieval speed of the system.
引文
[1] AMIRIAN P,BASIRI A. Landmark-Based Pedestrian Navigation Using Augmented Reality and Machine Learning[M]. Berlin:Springer International Publishing,2016.
[2]徐祥振.基于SIFT的地标导游软件的设计与实现[D].重庆:重庆大学,2017.
[3]罗文城.基于惯导数据和图像地标的室内定位关键技术研究[D].成都:电子科技大学,2018.
[4]孙亚丽.基于多传感器融合的室内地标识别系统[D].杭州:浙江理工大学,2018.
[5]范音.面向全景图的地标识别技术研究[D].武汉:华中科技大学,2015.
[6] LOWE D G. Object recognition from local scale-invariant features[C]//GAVRILA D M,PHILOMIN V. Proceedings of the Seventh IEEE International Conference on Computer Vision,Kerkyra:IEEE,1999:20-27.
[7] LOWE D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision,2004,60(2):91-110.
[8] BAY H,ESS A,TUYTELAARS T,et al. Speeded-up robust features(SURF)[J]. Computer Vision and Image Understanding,2008,110(3):346-359.
[9]徐政.基于图像检索的博物馆导览系统的研究和实现[D].广州:中山大学,2014.
[10] SEDAGHAT A,EBADI H. Remote sensing image matching based on adaptive binning SIFT descriptor[J]. IEEE Transactions on Geoscience and Remote Sensing,2015,53(10):5283-5293.
[11]林开司,张安东,邢兵锁.基于SURF算法的图像复制粘贴篡改检测[J].齐齐哈尔大学学报(自然科学版),2019,35(2):10-13.
[12]王卫兵,白小玲,徐倩. SURF和RANSAC的特征图像匹配[J].哈尔滨理工大学学报,2018,23(1):117-121.
[13]DEBBARMA S,SINGH A B,SINGH K M. Keypoints based copy-move forgery detection of digital images[C]//SRIHARI V,KALPANA P,ANITHA R,et al. 2014 International Conference on Informatics,Electronics&Vision,Dhaka:IEEE,2014:23-24.
[14]FISCHLER M A,BOLLES R C. Random sample consensus:a paradigm for model fitting with applications to image analysis and automated cartography[J]. Communications of the ACM,1981,24(6):381-395.
[15]赵明富,陈海军,宋涛,等.改进RANSAC-SIFT算法在图像匹配中的研究[J].激光杂志,2018,39(1):114-118.
[16]谷宗运,谭红春,殷云霞,等.基于SURF和改进的RANSAC算法的医学图像配准[J].中国医学影像学杂志,2014,22(6):470-475.
[2]徐祥振.基于SIFT的地标导游软件的设计与实现[D].重庆:重庆大学,2017.
[3]罗文城.基于惯导数据和图像地标的室内定位关键技术研究[D].成都:电子科技大学,2018.
[4]孙亚丽.基于多传感器融合的室内地标识别系统[D].杭州:浙江理工大学,2018.
[5]范音.面向全景图的地标识别技术研究[D].武汉:华中科技大学,2015.
[6] LOWE D G. Object recognition from local scale-invariant features[C]//GAVRILA D M,PHILOMIN V. Proceedings of the Seventh IEEE International Conference on Computer Vision,Kerkyra:IEEE,1999:20-27.
[7] LOWE D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision,2004,60(2):91-110.
[8] BAY H,ESS A,TUYTELAARS T,et al. Speeded-up robust features(SURF)[J]. Computer Vision and Image Understanding,2008,110(3):346-359.
[9]徐政.基于图像检索的博物馆导览系统的研究和实现[D].广州:中山大学,2014.
[10] SEDAGHAT A,EBADI H. Remote sensing image matching based on adaptive binning SIFT descriptor[J]. IEEE Transactions on Geoscience and Remote Sensing,2015,53(10):5283-5293.
[11]林开司,张安东,邢兵锁.基于SURF算法的图像复制粘贴篡改检测[J].齐齐哈尔大学学报(自然科学版),2019,35(2):10-13.
[12]王卫兵,白小玲,徐倩. SURF和RANSAC的特征图像匹配[J].哈尔滨理工大学学报,2018,23(1):117-121.
[13]DEBBARMA S,SINGH A B,SINGH K M. Keypoints based copy-move forgery detection of digital images[C]//SRIHARI V,KALPANA P,ANITHA R,et al. 2014 International Conference on Informatics,Electronics&Vision,Dhaka:IEEE,2014:23-24.
[14]FISCHLER M A,BOLLES R C. Random sample consensus:a paradigm for model fitting with applications to image analysis and automated cartography[J]. Communications of the ACM,1981,24(6):381-395.
[15]赵明富,陈海军,宋涛,等.改进RANSAC-SIFT算法在图像匹配中的研究[J].激光杂志,2018,39(1):114-118.
[16]谷宗运,谭红春,殷云霞,等.基于SURF和改进的RANSAC算法的医学图像配准[J].中国医学影像学杂志,2014,22(6):470-475.