Abstract
In this paper, a method of half-edge composite structure (HECS) extraction is proposed. First, an edge is divided into two half-edges, and then one half-edge is combined to form a HECS according to the condition. A HECS is an infrastructure composed of two half-edges and one endpoint. Matching means to calculate the distance between the HECS description vectors of two images and to get high-quality matching according to the conditions. To describe a more complete image structure, we propose a graph structure combination method based on a HECS. The feature set of the graph structure is formed by the association of the HECS, and the matching relationship of the graph structure set is mapped by the matching results of the HECS. Based on the image structure set and visual residual phenomenon, a motion matching model of video stream is constructed, which is called the visual residual model. For numerical estimation of natural images, we propose an evaluation method to investigate the feature stability of video images. The result shows that performance of the algorithm discussed in this paper has a good performance in the numerical estimation of continuous matching feature stability shown in the video stream.
Similar content being viewed by others
References
Acton ST, Rossi A (2008) Matching and retrieval of tattoo images: active contour CBIR and glocal image features. Image Anal Interpret. https://doi.org/10.1109/SSIAI.2008.4512275
Bay H, Ess A, Tuytelaars T et al (2008) Speeded-up robust features (SURF). Comput Vis Image Underst 110(3):346–359. https://doi.org/10.1007/11744023_32
Dimai A (1999) Unsupervised extraction of salient region-descriptors for contentbased image retrieval. Int Conf Image Anal Process. https://doi.org/10.1109/ICIAP.1999.797674
Fujimoto S, Hu Z, Chapuis R et al (2016) ORB-SLAM map initialization improvement using depth. IEEE Int Conf Image Process. https://doi.org/10.1109/ICIP.2016.7532359
Jadhav LH, Momin BF (2016) Detection and identification of unattended/removed objects in video surveillance. IEEE Int Conf Recent Trends Electron. https://doi.org/10.1109/RTEICT.2016.7808138
Kim D, Kim H, Shin J et al (2019) Real-time multiple pedestrian tracking based on object identification. In: IEEE 9th international conference on consumer electronics (ICCE-Berlin). pp 363–365. https://doi.org/10.1109/ICCE-Berlin47944.2019.8966205
Li K, Yao J, Xiaohu Lu, Li Li, Zhang Z (2016) Hierarchical line matching based on Line–Junction–Line structure descriptor and local homography estimation. Neurocomputing 184:207–220. https://doi.org/10.1016/j.neucom.2015.07.137
Li X, Li Y, Örnek EP, Lin J, Tombari F (2020) Co-planar parametrization for Stereo-SLAM and visual-inertial odometry. IEEE Robot Autom Lett 5(4):6972–6979. https://doi.org/10.1109/LRA.2020.3027230
Lowe DG (2004) Distinctive image features from scale-invariant keypoints. Int J Comput Vis 60(2):91–110. https://doi.org/10.1023/B:VISI.0000029664.99615.94
Rafael CG, Richard EW (2008) Digital image processing, 3rd edn. Pearson Education, pp 456–490
Rublee E, Rabaud V, Konolige K et al (2012) ORB: an efficient alternative to SIFT or SURF. Int Conf Comput Vis IEEE 2012:2564–2571. https://doi.org/10.1109/ICCV.2011.6126544
Saleem S, Sablatnig R (2014) A Robust SIFT descriptor for multispectral images. IEEE Signal Process Lett 21(4):400–403. https://doi.org/10.1109/LSP.2014.2304073
Shojanazeri H, Zhang D, Teng SW et al (2018) A novel perceptual dissimilarity measure for image retrieva. In: 2018 international conference on image and vision computing New Zealand (IVCNZ). pp 1–6. https://doi.org/10.1109/IVCNZ.2018.8634763
Souded M, Giulieri L, Bremond F (2011) An object tracking in particle filtering and data association framework, using SIFT features. Int Conf Imaging Crime Detect Prev. https://doi.org/10.1049/ic.2011.0104
Xiang Y, Wang F, You H (2018) OS-SIFT: a robust SIFT-like algorithm for high-resolution optical-to-SAR image registration in suburban areas. IEEE Trans Geosci Remote Sens 56(6):3078–3090. https://doi.org/10.1109/TGRS.2018.2790483
Xu L, Yan Q, Xia Y et al (2012) Structure extraction from texture via relative total variation. ACM Trans Graph 31(6):1. https://doi.org/10.1145/2366145.2366158
Xu K, Liu J, Miao J et al (2019) An improved SIFT algorithm based on adaptive fractional differential. J Ambient Intell Hum Comput 10:3297–3305. https://doi.org/10.1007/s12652-018-1055-1
Funding
Funding was provided by Chunhui Project Foundation of the Education Department of China (Z2014050), Sichuan Provincial Education Department of China (17ZA0360), Innovation Fund of Postgraduate, Xihua University (ycjj2017068)
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Qian, J., Luo, X. & Xue, Y. Half-edge composite structure: good performance in motion matching. J Ambient Intell Human Comput (2021). https://doi.org/10.1007/s12652-021-03073-4
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s12652-021-03073-4