Abstract
The existing shift technique for quaternion matrix computation is to use real shifts rather than quaternion shifts, because quaternions are multiplicatively non-commutative in general. This becomes the obstacle to the development of fast and stable algorithms for computing partial quaternion eigenpairs. To overcome this difficulty, a new inverse quaternion iteration with quaternion shift is derived for solving general right quaternion eigenvalue problem. Fixed and updated quaternion shift techniques are developed to speed up the convergence. The advantage of quaternion shifts is explained geometrically, as well as the fact that real shifts may lead to wrong results. The convergence theory is built in a different way to that in real and complex cases. Numerical experiments on simulated data and color face database demonstrate the efficiency and superiority of the newly proposed method over the state-of-the-art algorithms.
Similar content being viewed by others
Availability of supporting data
The datasets generated during and/or analyses during the current study are available in the website http://maths.jsnu.edu.cn/_t1395/5134/list.htm.
Notes
As far as we know, there is no convergence theory for real or complex inverse iteration with alterable shifts.
References
Chen, M.F., Chen, R.R.: Top eigenpairs of large scale matrices. CSIAM Trans. Appl. Math. 3(1), 1–25 (2022)
Chen, M.F., Jia, Z.G., Pang, H.K.: Computing top eigenpairs of Hermitizable matrix. Front. Math. China 16(6), 1–35 (2021)
Chen, M.F., Li, Y.S.: Development of powerful algorithm for maximal eigenpair. Front. Math. China 14(3), 493–519 (2019)
Hankins, T.L.: Sir William Rowan Hamilton. The Johns Hopkins University Press, Baltimore (1980)
He, Z.H., Wang, X.X., Zhao, Y.F.: Eigenvalues of quaternion tensors with applications to color video processing. J. Sci. Comput. 94, 1 (2023)
Huang, L., So, W.: On left eigenvalues of a quaternionic matrix. Linear Algebra Appl. 323, 105–116 (2001)
Jiang, T.: An algorithm for eigenvalues and eigenvectors of quaternion matrices in quaternionic quantum mechanics. J. Math. Phys. 45, 3334–3338 (2004)
Jiang, T.: Algebraic methods for diagonalization of a quaternion matrix in quaternionic quantum theory. J. Math. Phys. 46, 052106–052108 (2005)
Jia, Z.G.: The Eigenvalue Problem of Quaternion Matrix: Structure-Preserving Algorithms and Applications. Science Press, Beijing (2019)
Jia, Z.G., Jin, Q.Y., Ng, Michael K., Zhao, X.L.: Non-local robust quaternion matrix completion for large-scale color image and video inpainting. IEEE Trans. Image Process. 31, 3868–3883 (2022)
Jia, Z.G., Ng, Michael K.: Structure preserving quaternion generalized minimal residual method. SIAM J. Matrix Anal. Appl. 42(2), 616–634 (2021)
Jia, Z.G., Wei, M., Ling, S.T.: A new structure-preserving method for quaternion Hermitian eigenvalue problems. J. Comput. Appl. Math. 239, 12–24 (2013)
Jia, Z.G., Wei, M., Zhao, M.X., Chen, Y.: A new real structure-preserving quaternion QR algorithm. J. Comput. Appl. Math. 343, 26–48 (2018)
Krizhevsky, A., Sutskever, I., E.Hinton, G.: Imagenet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems, 25, (2012)
De Leo, S., Scolarici, G.: Right eigenvalue equation in quaternionic quantum mechanics. J. Phys. A 33, 2971–2995 (2000)
Lee, H.C.: Eigenvalues and canonical forms of matrices with quaternion coefficients. Proc. R. Irish Acad. Sect. A 52, 253–260 (1949)
Li, Y., Wei, M.S., Zhang, F.X., Zhao, J.L.: On the power method for quaternion right eigenvalue problem. J. Comput. Appl. Math. 345, 59–69 (2019)
Li, Y., Wei, M.S., Zhang, F.X., Zhao, J.L.: A real structure-preserving method for the quaternion LU decomposition, revisited. Calcolo 54, 1553–1563 (2017)
Wang, Q.W., Wang, X.X.: Arnoldi method for large quaternion right eigenvalue problem. J. Sci. Comput. 82, 58 (2020)
Wei, M.S., Zhang, F.X., Zhao, J.L.: Quaternion Matrix Computations. Nova Science Publishers, Hauppauge (2018)
Zhang, F.: Quaternions and matrices of quaternions. Linear Algebra Appl. 251, 21–57 (1997)
Zhang, F., Wei, Y.: Jordan canonical from of a partitioned complex matrix and its application to real quaternion matrices. Comm. Algebra 29(6), 2363–2375 (2001)
Acknowledgements
The authors are grateful to the handling editor and the anonymous referees for their useful comments and suggestions, which greatly improved the original presentation. The second author gave a report about this work on 2022 postgraduate academic innovation forum on “Scientific Computing and Its Application" in Jiangsu province of China. We thank Professor Wen-Wei Lin for impressive comments and helpful suggestions.
Funding
This paper is supported in part by National Natural Science Foundation of China under grants 12171210, 12090011, 11771188 and 11771189, Natural Science Research of Jiangsu Higher Education Institutions of China under grant 21KJA110001, and Natural Science Foundation of Fujian Province of China under grant 2020J05034.
Author information
Authors and Affiliations
Contributions
Z. Jia and Q. Wang completed the writing of this paper and implemented the numerical comparison; H.-K. Pang and M. Zhao provided the testing data and designed the numerical examples. H.-K. Pang, M. Zhao and Z. Jia contributed equally to this work. All authors reviewed the last version.
Corresponding author
Ethics declarations
Ethical Approval and Consent to participate
Not applicable.
Consent for publication
Not applicable.
Human and Animal Ethics
Not applicable.
Conflict of interest
The authors declare that they have no competing interests.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Quaternion Power Method and Inverse Quaternion Iteration with L\(_2\)-norm
New versions of quaternion power method and inverse quaternion iteration are proposed by substituting L\(_\infty \)-norm with L\(_2\)-norm. Their pseudo codes are given in Algorithms 5 and 6, respectively.
Shifted Inverse Quaternion Iteration with L\(_2\)-norm
We substitute L\(_{\infty }\)-norm with L\(_2\)-norm in the shifted inverse quaternion iterations (Algorithms 3 and 4). Their pseudo codes are shown in Algorithms 7 and 8. Notice that the implement of steps 5–7 in Algorithms 7 and 8 is not equal to line 2 of Algorithm 6. The reason is that we need not get the exact solution of quaternion linear system in the inner loop.
Applications
Now we apply the proposed shifted inverse quaternion iteration algorithms to color image reconstruction.
1.1 Subspace Computation with Deflation
Assume that we have achieved the first projection vector \({w}_1\) that is calculated by the inverse quaternion iteration with a quaternion shift. The second projection vector \({w}_2\) is computed in a similar way
It is worth to mention that the projection vector \({w}_2\) at each iteration have to be orthogonalized against the previous \({w}_1\) with a Gram-Schmidt procedure. We calculate r projection vectors by weights, and write \(\textbf{W}=[{w}_1, {w}_2, \cdots , {w}_r]\), \(1\le r <n\). After deflating the sample by r directions, it is natural to reach that
where \(j=1,2,\cdots ,r+1\), and \(\textbf{F}^{deflated}{w}_j\) are zero, \(j=1,2,\cdots ,r\). If not, the feature information may be lost due to the interference from other directions. Set \(j=r+1\), (3.2) suggests that the corresponding vector is not linearly represented, i.e., the freshly computed projector is not in the subspace generated by the projected vectors. Therefore, we must orthogonalize the computed projector against the known ones.
1.2 Color Image Reconstruction with Projection
Let \(\textbf{F}_1, \textbf{F}_2, \cdots , \textbf{F}_l\in \mathbb {Q}^{m\times n}\) be l training samples that are represented by quaternion matrices, whose mean is noted as
The eigenface subspace is defined by \(\textbf{W}=[{w}_1, {w}_2, \cdots ,\) \( {w}_r]\). With computed the projections of l training images in the subspace \(\textbf{W}\), all we need to do is to make the projection to the subjection to the subspace, i.e.,
In other words, we reconstruct the image only by the first r vectors. It should be noted here that there is no need to amplify the effect of features in the process of reconstruction, so the projection matrix \(\textbf{W}\) is unweighted. The process mentioned above is simplified as Algorithm 9.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Jia, Z., Wang, Q., Pang, HK. et al. Computing Partial Quaternion Eigenpairs with Quaternion Shift. J Sci Comput 97, 41 (2023). https://doi.org/10.1007/s10915-023-02355-7
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10915-023-02355-7