DOI 10.17586/0021-3454-2023-66-5-399-408
UDC 519.614
TWO-LEVEL AND MODULARLY TWO-LEVEL QUASI-ORTHOGONAL WALSH-STRUCTURED MATRICES FOR IMAGE MASKING
St. Petersburg State University of Aerospace Instrumentation, Department of Computer Systems and Networks ; Researcher
Reference for citation: Sergeev А. M. Two-level and modularly two-level quasi-orthogonal Walsh-structured matrices for image masking. Journal of Instrument Engineering. 2023. Vol. 66, N 5. P. 399—408 (in Russian). DOI: 10.17586/0021-3454-2023-66-5-399-408.
Abstract. A separate class of quasi-orthogonal matrices, namely, two-level Mersenne matrices structured according to Walsh, are studied. The difference between the systems of orthogonal Hadamard–Walsh and Mersenne–Walsh functions is shown. Modular two-level Mersenne–Walsh matrices and their portraits are considered. A system of functions constructed using a modularly two-level Mersenne–Walsh matrix has twice as many levels as a system of functions constructed on the basis of a two-level Mersenne matrix structured according to Walsh. As an applied problem using structured quasi-orthogonal matrices, the procedure for masking images with two-level and modularly two-level Mersenne-Walsh matrices with an assessment of the results of masking - destruction of the original image is considered. The example of a test image demonstrates the change in the brightness histogram and the influence of the order of the masking matrix on the masking result.
Abstract. A separate class of quasi-orthogonal matrices, namely, two-level Mersenne matrices structured according to Walsh, are studied. The difference between the systems of orthogonal Hadamard–Walsh and Mersenne–Walsh functions is shown. Modular two-level Mersenne–Walsh matrices and their portraits are considered. A system of functions constructed using a modularly two-level Mersenne–Walsh matrix has twice as many levels as a system of functions constructed on the basis of a two-level Mersenne matrix structured according to Walsh. As an applied problem using structured quasi-orthogonal matrices, the procedure for masking images with two-level and modularly two-level Mersenne-Walsh matrices with an assessment of the results of masking - destruction of the original image is considered. The example of a test image demonstrates the change in the brightness histogram and the influence of the order of the masking matrix on the masking result.
Keywords: quasi-orthogonal matrices, Hadamard–Walsh matrices, Mersenne–Walsh matrices, modular two-level matrices, image masking
References:
References:
- Horadam K.J. Hadamard matrices and their applications, Princeton University Press, 2007, 263 р.
- Seberry J., Yamada M. Hadamard Matrices: Constructions using number theory and linear algebra, Wiley, 2020, 384 p.
- Mironovsky L.A., Slaev V.A. Strip-Method for Image and Signal Transformation, Berlin, Boston, De Gruyter, 2011, https://doi.org/10.1515/9783110252569.
- Wang R. Introduction to Orthogonal Transforms with Applications in Data Processing and Analysis, Cambridge University Press, 2010, 504 p.
- Turyn R.J. Journal of Combinatorial Theory, Series A, 1974, vol. 16, рр. 313–333.
- Vostrikov А., Sergeev M. Smart Innovation, Systems and Technologies, 2015, vol. 40, рр. 161–168, DOI: 10.1007/978-3-319-19830-9_15.
- Seberry J. Bulletin of the Australian Mathematical Society, 1970, no. 2, pp. 45–54.
- Evangelaras H., Koukouvinos C., Seberry J. Journal of Telecommunications and Information Technology, 2003, vol. 2, рр. 3–10.
- Sergeev A.M., Vostrikov A.A. Spetsial'nyye matritsy: vychisleniye i primeneniye (Special Matrices: Calculation and Application), St. Petersburg, 2018, 112 р. (in Russ.)
- Balonin N.A., Sergeev M.B. Spetsial'nyye matritsy: psevdoobratnyye, ortogonal'nyye, adamarovy i kritskiye (Special Matrices: Pseudoinverse, Orthogonal, Hadamard and Cretan), St. Petersburg, 2019, 196 р., DOI: 10.25960/7325-1155-0.
- Sergeev A.M. Journal of Instrument Engineering, 2021, no. 2(64), pp. 90–96. (in Russ.)
- Jenny suggested this idea a year ago at an International meeting on Hadamard Matrices, http://mathscinet.ru/catalogue/files/Judy-AnneOsborne_JennysIdea_VisHadamard.pdf.
- Sergeev А., Sergeev M., Vostrikov A., Kurtyanik D. Smart Innovation, Systems and Technologies, 2019, vol. 143. рр. 135–143, DOI: 10.1007/978-981-13-8303-8_12.
- Balonin N.A., Sergeev M.B. Journal of Instrument Engineering, 2014, no. 10(57), pp. 38–41. (in Russ.)
- Vostrikov A., Sergeev M., Balonin N., Sergeev A. Procedia Computer Science, 2018, vol. 126, рр. 1054–1061, DOI: 10.1016/j.procS.2018.08.042.
- Rademacher H. Mathematische Annalen, 1922, no. 1–2(87), pp. 112–138.
- Harmuth H.F. IEEE Spectrum, 1969, vol. 6, рр. 82–91.
- Walsh J.L. American Journal of Mathematics, 1923, vol. 45, рр. 5–24.
- Balonin N.А., Vostrikov А.A., Sergeev M.В. Vestnik komp'iuternykh i informatsionnykh tekhnologii (Herald of Computer and Information Technologies), 2014, no. 11(125), pp. 51–56, DOI: 10.14489/vkit.2014.011.pp.051-056. (in Russ.)
- Mersenne-Walsh Matrices, http://mathscinet.ru/catalogue/walsh/.
- Kapranova E.A., Nenashev V.A., Sergeev M.B. Proceedings of SPIE – The International Society for Optical Engineering, "Image and Signal Processing for Remote Sensing XXIV", 2018, рр. 1078923, DOI:10.1117/12.2324249.
- Balonin N., Vostrikov A., Sergeev M. Smart Innovation, Systems and Technologies, 2015, vol. 40, рр. 141–147, DOI: 10.1007/978-3-319-19830-9_13.
- Sergeev A.M. Metody preobrazovaniya izobrazheniy i kodirovaniya signalov v kanalakh raspredelennykh sistem na osnove ispol'zovaniyaya spetsial'nykh kvaziortogonal'nykh matrits (Methods for Image Transformation and Signal Coding in Channels of Distributed Systems Based on the Use of Special Quasi-Orthogonal Matrices), Candidate’s Thesis, St. Petersburg, 2019, 153 р. (in Russ.)
- The Lenna Story, http://lenna.org/.