Skip to main content
SpringerLink
Log in

An eigenvector approach for obtaining scale and orientation invariant classification in convolutional neural networks

  • Original Article
  • Published:
Advances in Computational Intelligence Aims and scope Submit manuscript

Abstract

The convolution neural networks are well known for their efficiency in detecting and classifying objects once adequately trained. Though they address shift in-variance up to a limit, appreciable rotation and scale in-variances are not guaranteed by many of the existing CNN architectures, making them sensitive towards input image or feature map rotation and scale variations. Many attempts have been made in the past to acquire rotation and scale in-variances in CNNs. In this paper, an efficient approach is proposed for incorporating rotation and scale in-variances in CNN-based classifications, based on eigenvectors and eigenvalues of the image covariance matrix. Without demanding any training data augmentation or CNN architectural change, the proposed method, ‘Scale and Orientation Corrected Networks (SOCN)’, achieves better rotation and scale-invariant performances. SOCN proposes a scale and orientation correction step for images before baseline CNN training and testing. Being a generalized approach, SOCN can be combined with any baseline CNN to improve its rotational and scale in-variance performances. We demonstrate the proposed approach’s scale and orientation invariant classification ability with several real cases ranging from scale and orientation invariant character recognition to orientation invariant image classification, with different suitable baseline architectures. The proposed approach of SOCN, though is simple, outperforms the current state of the art scale and orientation invariant classifiers comparatively with minimal training and testing time.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15

Similar content being viewed by others

References

  • Banerjee B, Bhattacharjee T, Chowdhury N (2010) Image object classification using scaleinvariant feature transform descriptor with support vector machine classifier with histogramintersection kernel. In: International Conference on Advances in Information andCommunication Technologies, Springer, Berlin, Heidelberg, pp 443–448

  • Bay H, Tuytelaars T, Van Gool L (2006) Surf: speeded up robust features. In: Leonardis A, Bischof H, Pinz A (eds) Computer vision—ECCV 2006. Springer, Berlin, pp 404–417. ISBN:978-3-540-33833-8

  • Cheng G, Zhou P, Han J (2016) Rifd-cnn:Rotation-invariant and fisher discriminative convolutional neural networks for object detection.In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2884–2893

  • Ciregan D, Meier U, Schmidhuber J (2012) Multi-column deep neural networks for image classification.In: 2012 IEEE conference on computer vision and pattern recognition, IEEE, pp. 3642–3649

  • Ciregan D, Meier U, Schmidhuber J (2012) Multi-column deep neural networks for image classification. In: 2012 IEEE conference on computer vision and pattern recognition, pp 3642–3649. https://doi.org/10.1109/CVPR.2012.6248110

  • Cohen T, Welling M (2016) Group equivariant convolutional networks.In: International conference on machine learning, pp 2990–2999

  • Depeursinge A, Foncubierta-Rodríguez A, Van De Ville D, Müller H (2014) Rotation-covariant texture learning using steerable Riesz wavelets. IEEE Trans Image Process 23:898–908

    Article  MathSciNet  Google Scholar 

  • Dieleman S, De Fauw J, Kavukcuoglu K (2016) Exploiting cyclic symmetry in convolutional neural networks.In: International conference on machine learning, PMLR, pp 1889–1898

  • Esteves C, Allen-Blanchette C, Zhou X, Daniilidis K (2017) Polar transformer networks. arXiv preprint arXiv:1709.01889

  • Fasel B, Gatica-Perez D (2006) Rotation-invariant neoperceptron. In: 18th International Conference on Pattern Recognition (ICPR 06), IEEE Vol. 3, pp 336–339

  • Greenspan H, Belongie S, Goodman R, Perona P (1994) Rotation invariant texture recognition using a steerable pyramid. In: Proceedings of the 12th IAPR international conference on pattern recognition, vol 3—conference C: signal processing (Cat. No.94CH3440-5), Oct 1994, vol 2, pp 162–167. https://doi.org/10.1109/ICPR.1994.576896

  • Hamid NA, Sjarif NNA (2017) Handwritten recognition using SVM, KNN and neural network arXiv preprint arXiv:1702.00723

  • Huang FJ, LeCun Y (2006) Large-scale learning with SVM and convolutional nets for generic object categorization. In: Proceedings—2006 IEEE computer society conference on computer vision and pattern recognition, CVPR 2006, vol 1, pp 284–291. ISBN:0769525970. https://doi.org/10.1109/CVPR.2006.164

  • Jaderberg M, Simonyan K, Zisserman A, Kavukcuoglu K (2015) Spatial transformer networks. In: Cortes C, Lawrence ND, Lee DD, Sugiyama M, Garnett R (eds) Advances in neural information processing systems, vol 28, pp 2017–2025. Curran Associates, Inc., Red Hook. http://papers.nips.cc/paper/5854-spatial-transormer-networks.pdf

  • Jain A, Subrahmanyam GRKS, Mishra D (2017) Stacked features based CNN for rotation invariant digit classification. In: Shankar BU, Ghosh K, Mandal DP, Ray SS, Zhang D, Pal SK (eds) Pattern recognition and machine intelligence, pp 527–533. Springer International Publishing, Cham. ISBN:978-3-319-69900-4

  • Kandi H, Jain A, Chathoth SV, Mishra D, Subrahmanyam GRS (2019) Incorporating rotational invariance in convolutional neural network architecture. Pattern Anal Appl 22(3):935–948

  • Ke Y, Sukthankar R (2004) PCA-SIFT:A more distinctive representation for local image descriptors. In: Proceedings of the 2004 IEEE ComputerSociety Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004. Vol. 2, pp. II-II. IEEE

  • Khotanzad A, Hong YH (1990) Invariant image recognition by Zernike moments. IEEE Trans Pattern Anal Mach Intell 12(5):489–497. ISSN:0162-8828. https://doi.org/10.1109/34.55109

  • Kreutz M, Völpel B, Janßen H (1996) Scale-invariant image recognition based on higher-order autocorrelation features. Pattern Recognit 29:19–26. https://doi.org/10.1016/0031-3203(95)00078-X

    Article  Google Scholar 

  • Krizhevsky A, Sutskever I, Hinton GE (2012) ImageNet classification with deep convolutional neural networks. Neural Inf Process Syst 25:01. https://doi.org/10.1145/3065386

    Article  Google Scholar 

  • Laptev D, Savinov N, Buhmann JM, Pollefeys M (2016) Ti-pooling:transformation-invariant pooling for feature learning in convolutional neural networks.In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 289–297

  • Lecun Y, Bottou L, Bengio Y, Haffner P (1998) Gradient-based learning applied to document recognition. Proc IEEE 86(11):2278–2324. ISSN:0018-9219. https://doi.org/10.1109/5.726791

  • Li J, Yang Z, Liu H, Cai D (2017) Deep rotation equivariant network. Neurocomputing 290:26–33

    Article  Google Scholar 

  • Lin K, Lu J, Chen C, Zhou J, Sun M (2019) Unsupervised deep learning of compact binary descriptors. IEEE Trans Pattern Anal Mach Intell 41(6):1501–1514. ISSN:0162-8828. https://doi.org/10.1109/TPAMI.2018.2833865

  • Lowe DG (2004) Distinctive image features from scale-invariant keypoints. Int J Comput Vis 14(1):234–778

    Google Scholar 

  • Marcos D, Volpi M, Komodakis N, Tuia D (2017)Rotation equivariant vector field networks.In: Proceedings of the IEEE International Conference on Computer Vision, pp 5048–5057

  • Nam G, Choi H, Cho J, Kim I (2018) PSI-CNN: a pyramid-based scale-invariant CNN architecture for face recognition robust to various image resolutions. Appl Sci 8:1561. https://doi.org/10.3390/app8091561

    Article  Google Scholar 

  • Patel S (2018) A-z handwritten alphabets in .csv format. Available: https://www.kaggle.com/sachinpatel21/azhandwritten-alphabetsin-csv-format

  • Rifai S, Vincent P, Muller X, Glorot X, Bengio Y (2011) Contractive auto-encoders: Explicit invariance during feature extraction. In Icml

  • Ryu J, Yang MH, Lim J (2018) Dft-based transformation invariant pooling layer for visual classification. In: Proceedings of the European Conference on Computer Vision (ECCV), pp. 84–99

  • Rodriguez R, Dokladalova E, Dokládal (2019) Rotation invariant CNN using scattering transform for image classification.In: 2019 IEEE International Conference on Image Processing (ICIP), pp. 654–658. IEEE

  • Santosh KC (2011) Character recognition based on DTW-radon. In: International conference on document analysis and recognition, ICDAR, Sept 2011, pp 264–268. https://doi.org/10.1109/ICDAR.2011.61

  • Satoh S, Kuroiwa J, Aso H, Miyake S (1997) Recognition of rotated patterns using neocognitron. In: ICONIP

  • Scherer D, Behnke S (2010) Evaluation of pooling operations in convolutional architectures for object recognition. In: 20th International conference on artificial neural networks (ICANN)

  • Shang Y, An T, Meng Z (2010) A study on methods for rotation invariant image recognition based on texture characteristic. In: 2010 International conference on computer application and system modeling (ICCASM 2010), Oct 2010, vol 9, pp V9-568–V9-571. https://doi.org/10.1109/ICCASM.2010.5622967

  • Shen X, Tian X, He A, Sun S, Tao D (2016) Transform-invariant convolutional neural networks for image classification and search. In: ACM multimedia

  • Simonyan K, Zisserman A (2015) Very deep convolutional networks for large-scale image recognition. In: International conference on learning representations

  • Takahashi R, Matsubara T, Uehara K (2017) Scale-invariant recognition by weight-shared CNNs in parallel. In: Zhang M-L, Noh Y-K (eds) Proceedings of the ninth Asian conference on machine learning. Proceedings of machine learning research. PMLR, vol 77, 15–17 Nov 2017, pp 295–310. http://proceedings.mlr.press/v77/takahashi17a.html

  • Vincent P, Larochelle H, Lajoie I, Bengio Y, Manzagol P-A (2010) Stacked denoising autoencoders: learning useful representations in a deep network with a local denoising criterion. J Mach Learn Res 11:3371–3408

    MathSciNet  MATH  Google Scholar 

  • Worrall DE, Garbin SJ, Turmukhambetov D, Brostow GJ (2016) Harmonic networks: deep translation and rotation equivariance. In: 2017 IEEE conference on computer vision and pattern recognition (CVPR), pp 7168–7177

  • Wu W, Wei S (1996) Rotation and gray-scale transform-invariant texture classification using spiral resampling, subband decomposition, and hidden Markov model. IEEE Trans Image Process 5(10):1423–1434. ISSN:1057-7149. https://doi.org/10.1109/83.536891

  • Xiao H, Rasul K, Vollgraf R (2017) Fashion-MNIST: a novel image dataset for benchmarking machine learning algorithms. arXiv preprint arXiv:1708.07747

  • Xu Y, Xiao T, Zhang J, Yang K, Zhang Z (2014) Scale-invariant convolutional neural networks

  • Xu W, Wang G, Sullivan A, Zhang Z (2020) Towards learning affine-invariant representations via data-efficient CNNs. In: Proceedings of the IEEE/CVF winter conference on applications of computer vision (WACV)

  • Zhang H, Li Z, Liu Y (2016) Fractional orthogonal Fourier–Mellin moments for pattern recognition. In: Tan T, Li X, Chen X, Zhou J, Yang J, Cheng H (eds) Pattern recognition. CCPR. Communications in computer and information science, vol 662, pp 766–778. Springer, Singapore. ISBN:978-981-10-3002-4

  • Zhou Y, Ye Q, Qiu Q, Jiao J (2017) Oriented response networks. In: The IEEE conference on computer vision and pattern recognition (CVPR), July 2017

Download references

Author information

Authors and Affiliations

  1. IIST Thiruvananthapuram, Thiruvananthapuram, India

    Swetha Velluva Chathoth, Asish Kumar Mishra, Deepak Mishra & Subrahmanyam Gorthi R. K. Sai

Authors
  1. Swetha Velluva Chathoth
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Asish Kumar Mishra
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Deepak Mishra
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Subrahmanyam Gorthi R. K. Sai
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Deepak Mishra.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Chathoth, S.V., Mishra, A.K., Mishra, D. et al. An eigenvector approach for obtaining scale and orientation invariant classification in convolutional neural networks. Adv. in Comp. Int. 2, 8 (2022). https://doi.org/10.1007/s43674-021-00023-7

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s43674-021-00023-7

Keywords

Search

Navigation