Abstract
The region proposal algorithm for line insulators is a critical step in line insulator detection. The selective search algorithm provides a rich candidate area for line insulators. However, the selective search algorithm uses SIFT descriptor to extract texture features. It causes a large time overhead. In response to this problem, this paper proposes an improved algorithm. The algorithm use the Harr-Like algorithm to form a feature value map. Then, the formed feature value map is mapped to a fixed-size texture vector HBSN (Harr-Like Based on SPP-Net) by using Spatial Pyramid Pooling Network. In the process of the formation of feature maps and the construction of feature vectors, the integral image is used for acceleration. The integral image can only traverses the pixels in the original image once, forming an integral image of the feature values corresponding to the original image. When calculating the texture vector of different regions in the original image, it only needs to index the corresponding position of the eigenvalue integral image, which overcomes the problem of repeated calculation of texture features in the overlapping region by selective search and achieves fast calculation of texture similarity in different regions.
Experiments show that the improved selective search algorithm for line insulator has a calculation speed increase of 8.21% compared with the selective search.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Huang, K., Ren, W.: Overview of image object classification and detection algorithms. J. Comput. Sci. 1225–1240 (2014)
Uijlings, J.R.R., et al.: Selective search for object recognition. Int. J. Comput. Vis. 1(1), 154–171 (2013)
Girshick, R.: Fast R-CNN. In: 2015 IEEE International Conference on Computer Vision (ICCV), pp. 1440–1448. IEEE, Santiago (2015)
Ren, S., He, K., Girshick, R., et al.: Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans. Pattern Anal. Mach. Intell. 39(6), 1137–1149 (2015)
Li, X., Zhou, Z., Lu, S.: Face component detection based on selective search algorithms. Comput. Eng. Sci. 34(09), 1829–1836 (2018)
Naiel, M.A., Ahmad, M.O., Swamy, M.N.S.: A vehicle detection scheme based on two-dimensional HOG features in the DFT and DCT domains. Multidimensional Syst. Signal Process. 30(4), 1697–1729 (2018). https://doi.org/10.1007/s11045-018-0621-1
Song, D., Zhao, B.-J.: The object recognition algorithm based on affine histogram of oriented gradient. J. Electron. Inf. Technol. 35(6), 1428–1434 (2013)
Chen, Z., Yao, S., Liu, C., et al.: Sketch face recognition: P-HOG multi-features fusion. Int. J. Pattern Recogn. Artif. Intell. 33(04), 1–17 (2019)
Zhou, H., Yuan, Y., Shi, C.: Object tracking using SIFT features and mean shift. Comput. Vis. Image Understand. 113(3), 345–352 (2009)
Lv, G., Teng, S.W., Lu, G.: Enhancing SIFT-based image registration performance by building and selecting highly discriminating descriptors. Pattern Recogn. Lett. 84(34), 156–162 (2016)
Wu, S., Zhan, Y.: Face detection based on selective search and convolutional neural network. Comput. Appl. Res. 34(09), 2854–2876 (2017)
Felzenszwalb, P.F., Huttenlocher, D.P.: Efficient graph-based image segmentation. Int. J. Comput. Vis. 59(2), 167–181 (2004)
He, K., et al.: Spatial pyramid pooling in deep convolutional networks for visual recognition. IEEE Trans. Pattern Anal. Mach. Intell. 37(09), 1904–1916 (2015)
Yan, Z., Yang, F., Wang, J., Shi, Y., Li, C., Sun, M.: Face orientation detection in video stream based on Harr-like feature and LQV classifier for civil video surveillance. In: IET International Conference on Smart and Sustainable City 2013 (ICSSC 2013), pp. 161–165. IEEE, Shanghai (2013)
Wei, Y., et al.: Multi-vehicle detection algorithm through combining Harr and HOG features. Math. Comput. Simul. 155(12), 130–145 (2019)
Acknowledgement
This research is partially supported by:
1. Research Foundation of Education Bureau of Jilin Province (JJKN20190710KJ).
2. Science and Technology Innovation Development Plan Project of Jilin city (20190302202).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Guo, S., Yue, B., Bai, Q., Lin, H., Zhou, X. (2020). Region Proposal for Line Insulator Based on the Improved Selective Search Algorithm. In: Sun, X., Wang, J., Bertino, E. (eds) Artificial Intelligence and Security. ICAIS 2020. Lecture Notes in Computer Science(), vol 12240. Springer, Cham. https://doi.org/10.1007/978-3-030-57881-7_60
Download citation
DOI: https://doi.org/10.1007/978-3-030-57881-7_60
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-57880-0
Online ISBN: 978-3-030-57881-7
eBook Packages: Computer ScienceComputer Science (R0)