An Improved Indoor Image Registration Algorithm Based on Shallow Convolutional Neural Network Descriptor

  • Yun Gong
  • Mengjia YangEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11902)


At present, the application demand of indoor simultaneous localization and mapping (SLAM) technology increases greatly, among which, image matching is the most basic and critical content. Compared with traditional image registration, indoor image registration has higher requirements on the real-time and robustness of the algorithm. The shallow convolutional neural network is a deep machine learning model based on supervised learning with the characteristics of centralized and automatic learning from data. Aiming at the problems of slow processing and strong rotation failure of feature descriptors in traditional registration algorithms, this paper proposed an improved algorithm of local feature descriptor of triple-sample shallow convolutional neural network, which has strong feature expression ability. In addition, the performance of our improved algorithm was compared with that of three traditional algorithms (SIFT, ORB and SURF) in rotation change of indoor image matching. The results show that the improved algorithm performs better than the other three traditional methods and has a certain antagonistic effect on image rotation.


Feature detection Feature matching Neural network Descriptor 


  1. 1.
    Chai, H.: Data association method for mobile robots in SLAM. Doctor, Dalian University of Technology (2010)Google Scholar
  2. 2.
    Gao, X., Zhang, T., Liu, Y.: Visual SLAM Fourteen Lectures –Form Theory to Practice, 1st edn. China Machine Press, Beijing (2017)Google Scholar
  3. 3.
    Yan, K., Sukthankar, R.: PCA-SIFT: a more distinctive representation for local image descriptors. IEEE Comput. Soc. 2(2), 506–513 (2004)Google Scholar
  4. 4.
    Mikolajczyk, K., Schmid, C.: A performance evaluation of local descriptors. In: Computer Vision and Pattern Recognition, USA, pp. 257–263 (2003)Google Scholar
  5. 5.
    Tola, E., Lepetit, V., Fua, P.: DAISY: an efficient dense descriptor applied to wide-baseline stereo. IEEE Trans. Software Eng. 32(5), 815–830 (2010)Google Scholar
  6. 6.
    Geng, Z., Zhang, B., Fan, D.: Digital Photogrammetry. Surveying and Mapping Publishing House, Beijing (2010)Google Scholar
  7. 7.
    Leutenegger, S., Chli, M., Siegwart, R.Y.: BRISK: binary robust invariant scalable keypoints. In: IEEE International Conference on Computer Vision, Spain, pp. 2548–2555 (2011)Google Scholar
  8. 8.
    Vandergheynst, P., Ortiz, R., Alahi, A.: FREAK: fast retina keypoint. In: IEEE Conference on Computer Vision & Pattern Recognition, USA, pp. 510–571 (2012)Google Scholar
  9. 9.
    Yi, K.M., Trulls, E., Lepetit, V., Fua, P.: LIFT: learned invariant feature transform. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9910, pp. 467–483. Springer, Cham (2016). Scholar
  10. 10.
    Morevec, H.P.: Towards automatic visual obstacle avoidance. In: International Joint Conference on Artificial Intelligence, p. 584 (1977)Google Scholar
  11. 11.
    Simard, P.Y., Haffner, P., Lecun, Y.: Boxlets: a fast convolution algorithm for signal processing and neural networks. In: Conference on Advances in Neural Information Processing Systems, USA, pp. 571–577 (1999)Google Scholar
  12. 12.
    Hu, J.: Research on high-resolution aerial image feature matching technology. Doctor, East China University of Technology (2018)Google Scholar
  13. 13.
    Rosin, B.P.L.: Measuring corner properties. Comput. Vis. Image Underst. 73(2), 291–307 (1999)CrossRefGoogle Scholar
  14. 14.
    Fischer, P., Dosovitskiy, A., Brox, T.: Descriptor matching with convolutional neural networks: a comparison to SIFT. Comput. Sci. (2014)Google Scholar
  15. 15.
    Vijay, K.B.G., Carneiro, G., Reid, I.: Learning local image descriptors with deep siamese and triplet convolutional networks by minimizing global loss functions. In: IEEE Conference on Computer Vision & Pattern Recognition, USA, pp. 5385–5394 (2016)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Xi’an University of Science and TechnologyXi’anChina

Personalised recommendations