Digital
Signal Processing |
Scientific
& Technical Journal |

3
Keywords: information technology, Fourier processing, finite signal, parametric discrete Fourier transform, two-dimensional discrete Fourier transform with variable parameters, separability of the transform kernel.
The complication of problems solved by digital Fourier processing in information technology has stimulated the transition from one-dimensional to two-dimensional digital Fourier processing. A systematic analysis of the transition from the one-dimensional discrete Fourier transform (DFT) to the two-dimensional discrete Fourier transform (2D DFT) showed that, firstly, such a transition is far from trivial and, secondly, the transition is primarily qualitative, not quantitative character. At the same time, the generalization of the results of the two-dimensional case to the multidimensional one, as a rule, does not cause difficulties, since it is mainly quantitative, not qualitative. As is known, for the practical application of Fourier processing methods, expanding the scope of their application, an important role belongs to the procedures for the rapid implementation of the corresponding Fourier transforms. The story of the FFT algorithm, proposed in 1965, is a vivid confirmation of this. The article deals with the solution of an important and urgent problem of developing fast algorithms for implementing a new discrete Fourier transform: a two-dimensional discrete Fourier transform with variable parameters (2D DFT-VP). In this paper, the following three groups of methods for improving the speed of 2D DFT-VP are proposed and studied. The first group of methods for improving the speed of 2D DFT-VP is based on the separability property of the core of 2D DFT-VP and the use of one-dimensional parametric DFTs (DFT-P). The second group of methods for improving the performance of 2D DFT-VP is based on the separability property of the 2D DFT-VP kernel and the use of one-dimensional parametric fast Fourier transforms (1D FFT-P). The third group of methods for improving the performance of 2D DFT-VP is based on the 2D Fast Fourier Transform (2D FFT-VP) in vector base 2, with space decimation. A comparative analysis of the effectiveness of the proposed three groups of methods for improving the speed of 2D DFT-VP based on a computer complex developed in the fourth generation language MATLAB (4GLS) was carried out. 2. Favorskaya M., Savchina E., Popov A. Adaptive visible image watermarking based on Hadamard transform, IOP Conference Series: Materials Science and Engineering, 2018, vol. 450, no. 5, pp. 052003.1-052003.6. doi: 10.1088/1757-899X/450/5/052003 3. Klionskiy D. M., Kaplun D. I., Geppener V. V. Empirical more decomposition for signal preprocessing and classification of intrinsic mode functions, Pattern Recognition and Image Analysis (Advances in Mathematical Theory and Applications), 2018, vol. 28, no. 1, pp. 122–132. doi:10.1134/S1054661818010091 4. Ponomarev A. V., Ponomareva O. V. Digital technologies in non-destructif testing, Journal of Physics: Conference Series, 2019, pp. 12038. 5. Ponomareva O.V., Ponomarev A.V. Application of parametric discret Fourier transform non-destructif testing of composite materiaials with a free oscilation metod, Journal of Physics: Conference Series, 2019, pp. 12039. 6. Batishchev V. I.,Volkov I. I., Zolin A. G. Using a stochastic basis in signal and image recovery problems, Optoelectronics, Instrumentation and Data Processing, 2017, vol. 53, no. 4, pp. 414–420. 7. Kulikovskikh I., Prokhorov S. Psychological perspectives on implicit regularization: a model of retrieval-induced forgetting (RIF), Journal of Physics: Conference Series. electronic edition, 2018, pp. 012079, doi:10.1088/1742-6596/1096/1/012079 8. Favorskaya M. N., Buryachenko V. V. Authentication and copyright protection of videos under transmitting specifications, Computer Vision in Advanced Control Systems-5. ISRL, Springer, Cham, 2020, vol. 175, pp. 119–160, doi.org/10.1007/978-3-030-33795-7_5 9. Blahut R. E. Fast Algorithms for Digital Signal Processing, Reading, MA, Addison-Wesley, 1984. 10. Bogner R.E., Constantinides A.G. Introduction to digital filtering. John Wiley and Sonc, London, New York, Sydney, Toronto, 1975. 216 p. 11. Likhttsinder B. Conditional average value of queues in queuing systems with bath request flows, 2017 4th International Scientific-Practical Conference Problems of Infocommunications Science and Technology, PIC S and T 2017 – Proceedings, 2018, pp. 49-52, doi:10.1109/INFOCOMMST.2017.8246347 12. Bakulin M.G., Vityazev V.V., Shumov A.P., Kreyndelin V.B. Effective signal detection for the spatial multiplexing mimo systems, Telecommunications and Radio Engineering, 2018, vol. 77, no. 13, pp. 1141–1158. doi.org/10.1615/TelecomRadEng.v77.i13.30. 13. Prozorov D., Tatarinova A. Comparison of grapheme-to-phoneme conversions for spoken document, 2019 IEEE East-West Design and Test Symposium, EWDTS 2019, 2019, pp. 8884449, doi:10.1109/EWDTS.2019.8884449 14. Prozorov D., Trubin I. Detection of a signal in the simo system with spatial correlation of noise, 2018 7th Mediterranean Conference on Embedded Computing, MECO 2018 - Including ECYPS 2018, 2018, pp. 1–5, doi:10.1109/MECO.2018.8405965 15. Urakov A., Gurevich K., Alies M., Reshetnikov A., Kasatkin A., Urakova N. The tissue temperature during injection of drug solution into it as an integral indicator of rheology, Journal of Physics: Conference Series. 4th International Conference on Rheology and Modeling of Materials, IC-RMM 2019, 2020, pp. 012003, doi:10.1088/1742-6596/1527/1/012003 16. Gonzalez R.C., Woods R.E. Digital Image Processing, Published by Pearson, 2018, 1168 p. 17. Pratt William K. Digital image processing, A Wiley-Interscience publication 2007, 807 p. 18. Cooley J., Tukey J. An Algorithm for the Machine Calculation of Complex Fourier Series, Math. Comput., vol. 19, no. 90, Apr. 1965, pp. 297–301, doi: 10.2307/2003354. 19. Ponomarev A. V. Fundamentals of the theory of two-dimensional digital signal processing in Fourier bases with variable parameters, Digital signal processing, 2019, no. 2, pp. 12–20 (in Russian). 20. Ponomareva O. V., Ponomarev A. V. Interpolation in the spatial domain of 2D discrete signals using fast Fourier transforms, Intellektual'nye sistemy v proizvodstve, 2019, vol. 17, no. 1, pp. 88-94, doi: 10.22213/24-10-9304-2019-1-88-94 (in Russian). 21. Ponomareva O. V., Ponomarev A. V., Smirnova N. V. Digitalization of measurements of spectra in the Fourier basis - development trends and problems, Devices and Methods of Measurements, 2019, vol. 10, no. 3, pp. 271–280, doi: 10.21122/2220-9506-2019-10-3-271-280 (in Russian). 22. Ponomareva O. V., Ponomarev V. A. Measurement of the current energy Fourier spectrum of complex and real discrete signals at finite intervals, Intellectual systems in production, 2013, no.2 (22), pp. 149–157 (in Russian.). 23. Ponomareva O. V. Probabilistic properties of spectral estimates obtained by the parametric discrete Fourier transform, Intellektual'nye sistemy v proizvodstve, 2010, no. 2(16), pp. 36–42 (in Russian). 24. Dzhenkins G., Vatts D. Spectral analysis and its applications, vol.1, Moscow, Mir, 1971, 312 p. (in Russian). 25. Dzhenkins G., Vatts D. Spectral analysis and its applications, vol.2, Moscow, Mir, 1971, 287 p. (in Russian). 26. Milent'ev V. S., Batishchev V. I. Approximation methods and systems for measuring and monitoring parameters of periodic signals, Moscow, Fizmatlit, 2011, 240 p. (in Russian). 27. Bendat Dzh., Pirsol A. Applied analysis of random processes: Trans. from English, Moscow, Mir, 1989, 540 p. (in Russian). 28. Marpl-ml. S. L. Digital Spectral Analysis and its Applications, Moscow, World, 1990, 584 p. (in Russian). 29. Oppengejm E. H. The use of digital signal processing, Moscow, Mir, 1980, 552 p. (in Russian). 30. Yaglom A. M. Correlation theory of stationary random functions with examples from meteorology, Gidrometeoizdat. Leningrad, 1981, 281 p. (in Russian). 31. Batishchev V. I., Zolin A. G.,Kosarev D. N., Romaneev A. E. Approximation approach to solving the problems of analyzing and interpreting experimental data, Herald of Samara State University. Series: Engineering, 2006, no. 40, pp. 57–65 (in Russian). 32. Batishchev V. I., Melent'ev V. S. Measuring and modeling approach to determining the integral characteristics of periodic signals, News of higher educational institutions. Electromechanics, 2003, no.6, pp.36–39 (in Russian). 33. Batishchev V. I., Volkov I. I., Zolin A. G. The use of the stochastic basis in the problems of the restoration of signals and images, Avtometriya, 2017, vol. 53, no.4, pp.127–134 (in Russian). 34. Batishchev V. I., Volkov I. I., Zolin A. G. The study of the approximation properties of functional bases in the tasks of image reconstruction during remote sensing of the earth, Control and modeling problems in complex systems, works of the XVIII International Conference, 2016, pp. 304–307 (in Russian). 35. Prokhorov S. A., Kulikovskikh I. M. Unique Condition for generalized Laguerre Functions to solve pole Position Problem, Signal Processing, 2015, vol. 108, pp. 25–29. 36. Prohorov S. A., Grafkin V. V. Structural and spectral analysis of random processes, Samarskij nauchnyj centr RAN. Samara, 2010. (in Russian). 37. Prozorov D. E., Petrov E. P. Quick search for noise-like signals, O-kratnoe. Kirov, 2006 (in Russian). 38. Ponomareva O. V. Development of the theory and development of methods and algorithms for digital processing of information signals in parametric Fourier bases, Dissertation of the doctor of technical sciences, Izhevsk, 2016, 357 p. (in Russian). 39. Dudgeon D,E., Mersereau R.M. Multidimensional digital signal processing, Prentice-Hall, Inc., Englewood Cliffs, 1984. 488 p. 40. Ponomareva O.V., Ponomarev A.V. [Interpolation in the spatial domain of two-dimensional discrete signals using fast Fourier transforms] Intellektual'nye sistemy v proizvodstve, 2019, Vol 17, no. 1, pp. 88-94. (in Russ.) DOI 10.22213 / 2410-9304-2019-88-94 41. Ponomarev V.A., Ponomareva O.V. [Trends in the development of discrete indirect measurements of the parameters of electrical signals]. Metrology, 2017, no.1, pp.20-32 (in Russ.). 42. Ponomareva O.V., Ponomarev A.V., Smirnova N.V. [Digitalization of spectral measurements in the Fourier basis - development trends and problems]. Instruments and methods of measurements. 2019, no. 3, pp. 271-280. DOI: 21122 / 2220-95062-2019-10-3-271-280 43. Ponomareva O.V., Ponomarev A.V., Smirnova N.V. [Algorithms for Direct and Inverse Parametric Fast Fourier Transform]. Informacionnye tehnologii. 2022, no. 1, pp. 9-19. DOI: 21122 / it.28. 44. Ponomareva O.V., Ponomarev A.V. THEORETICAL FOUNDATIONS OF DIGITAL VECTOR FOURIER ANALYSIS OF TWO-DIMENSIONAL SIGNALS PADDED WITH ZERO SAMPLES Information and Control Systems. 2021. ¹ 1 (110). Ñ. 55-64. 45. Ponomarev A.V., Ponomareva O.V. [THE ROLE OF ZERO PADDING IN THE THEORY OF TWO-DIMENSIONAL FOURIER SIGNAL PROCESSING]. Digital signal processing,2022, no. 3, pp. 50-55. (in Russ.)
2. Syuzev V. V. Osnovy teorii cifrovoj obrabotki signalov. Moscow: RT Soft, 2014. – 752 p. (in Russian). 3. Kuo S. M., Gan W.-S. Digital signal processors: architectures, implementations and applications. Prentice Hal, 2004. 624 p. 4. Woods R., McAllister J., Lightbody G., Ying Yi. FPGA-based implementation of signal processing systems, 2-nd ed. Willey, 2017. 360 p. 5. Welch T. B. , Wright H. G., Morrow M.G. Real-time digital signal processing from MATLAB to C with the TMS320C6x DSPs, 3rd ed. CRC Press, 2017. 480 p. 6. Vityazev S.V. Cifrovyk processory obrabotki signalov (Difital signal processors). M.: Goryachaya liniya-Telkom, 2017. 100 s. 7. Farhang-Boroujeny B. Adaptive filters theory and applications, 2-nd ed. John Wiley & Sons, 2013. 778 p. 8. Djigan V.I. Adaptivnaya fil'traciya signalov: teoriya i algoritmy (Adaptive signal filtering: theory and algorithms). M: Tekhnosfera, 2013. 528 s. (In Russian). 9. Haykin S. Adaptive filter theory, 5th ed. Pearson Education Inc., 2014. 889 p. 10. Diniz P. S. R. Adaptive filtering algorithms and practical implementation, 5-th ed. Springer, 2020. 495 p. 11. Benesty J., Huang Y., Eds. Adaptive signal processing: applications to real-world problems. Sprringer-Verlag, 2003. 356 p. 12. Gay S. L., Benesty J., Eds. Acoustic signal processing for telecommunications. Springer, 2000. 333 p. 13. Vaseghi S. V. Advanced digital signal processing and noise reduction, 2-nd ed. John Wiley & Sons Ltd., 2000. 473 p. 14. Davis G. M. Noise reduction in speech applications. CRC Press, 2002. 432 p. 15. Makino S. Acoustic echo cancellation // IEEE Signal Processing Magazine. 1997. V. 14. ¹ 5. P. 39–41. 16. Albu I. , Anghel C., Paleologu C. Adaptive filtering in acoustic echo cancellation systems – a practical overview // Proceedings of the 9-th International Conference on Electronics, Computers and Artificial Intelligence (ECAI). Targoviste, Romania, June 29 – July 1, 2017. 6 p. 17. Benesty J., Chen J., Huang Y. Microphone array signal processing. Springer, 2008. 250 p. 18. Bai M. R., Ih J.-G., Benesty J. Acoustic array systems. Theory, implementation and application. John Wiley & Sons, 2013. 516 p. 19. Elliot S. J., Nelson P. A. Active noise control // IEEE Signal Processing Magazine. 1993. V. 10. ¹ 4. P. 12–35. 20. Kuo S. M., Morgan D. S. Active noise control: a tutotial review // Proceedings of the IEEE. 1999. V. 87. ¹ 6. P. 943–973. 21. Jariwala R., Upadhyaya I., George N. V. Robust equalizer design for adaptive room impulse response compensation // Applied Acoustics. 2017. Vl. 125. P. 1–6. 22. Cecchi S., Carini A., Spors S. Room response equalization – a review // Applied Sciences. 2018. V. 8, ¹ 1 (16). P. 1–47. 23. Qureshi S. Adaptive equalization // Proceedings of the IEEE. 1985. Vl. 73. ¹ 9. P. 1349–1387. 24. Djigan V.I. Ekvalajzery s drobnoj zaderzhkoj i obratnoj svyaz'yu na baze bystryh RLS-algoritmov (Fractionally spaced feed-backward equalizers, based on fast RLS adaptive filtering algorithms) // Problemy razrabotki perspektivnyh mikro- i nanoelektronnyh sistem (MES) (Problems of Perspective Micro- and Nanoelectronic Systems Development). 2020. Vypusk 2. S. 126–131. (In Russian). 25. Nezami M. K. Fundamentals of power amplifier linearization using digital pre-distortion // High Frequency Electronics. 2004. V. 3. ¹ 8. P. 54–59. 26. Gharaibeh K. M. Nonlinear distortion in wireless systems: modeling and simulation with MATLAB. John Wiley & Sons Ltd., 2012. 355 p. 27. Widrow B., Stearns D. D. Adaptive signal processing. Pearson. 1985. 528 p. 28. Giordano A. A., Hsu F. M. Least square estimation with application to digital signal processing. John Wiley & Sons, 1985. 412 p. 29. Djigan V.I. Mnogokanal'nye RLS- i bystrye RLS-algoritmy adaptivnoj fil'tracii (Multichannel RLS and fast RLS adaptive filtering algorithms ) // Uspekhi sovremennoj radioelektroniki (Successes of Modern Radioelectronics). 2004. ¹ 11. S. 48–77. (In Russian). 30. Djigan V. I. Recursive least squares – an idea whose time has come // Proceedings of the 7-th International Workshop on Spectral Methods and Multirate Signal Processing. Moscow, Russia, September 1 – 2, 2007. 4 p. 31. Bjarnason E. Active noise cancellation using a modified form of the filtered-X LMS algorithm // Proceedings of the 6-th European Signal Processing Conference. Brussels, Belgium, August 24 – 27, 1992. P. 1053–1056. 32. Bouchard M., Quednau S. Multichannel recursive-least-squares algorithms and fast-transversal-filter algorithms for active noise control and sound reproducing system // IEEE Transactions on Speech and Audio Processing. 2000. V. 8. ¹ 5. P. 606–618. 33. Douglas S. C. An efficient implementation of the modified filtered-X LMS algorithm // IEEE Signal Processing Letter. 1997. V. 4, ¹ 10. P. 286–288. 34. Akhtar M. T., Abe M., Kawamata M. Modified-filtered-x LMS algorithm based active noise control systems with improved online secondary-path modeling //Proceedings of the 47-th IEEE International Midwest Symposium on Circuits and Systems. Hiroshima, Japan, July 25 – 28, 2004. P. 1.13–1.16. 35. Poltman R. D. Conversion of the delayed LMS algorithm into the LMS algorithm // IEEE Signal Processing Letters. 1995. Vl. 2. ¹ 12. P. 223. 36. Djigan V. I. Petrovsky A. A., Qin Jingfan, Song Yang. Modified hybrid active noise control system // Proceedings of the 13-th IEEE East-West Design & Test Symposium (EWDTS-2015). Batumi, Georgia, September 26 – 29, 2015. P. 82–86. 37. Djigan V. I. Petrovsky A. A., Qin Jingfan, Song Yang. Active noise cancellation device // United State Patent ¹ US 10174411 B2, G10 11/178, H04R 1/10. Publ. 04.12.2018. 44 p. 38. Djigan V. I., Dolgikh D. A., Smekalov A. I., Shi H. Predistortion device // United State Patent ¹ US 10171041 B2, H 03F 1/32. Publ. 01.01.2019. 36 p. 39. Djigan V.I. Prediskaziteli signalov s pryamym obucheniem dlya usilitelej moshchnosti (Direct learning digital predistorters for power amplifiers) // Problemy razrabotki perspektivnyh mikro- i nanoelektronnyh sistem (MES) (Problems of Perspective Micro- and Nanoelectronic Systems Development). 2020. Vypusk 3. S. 151–157. (In Russian). 40. Allen J. B., Berkley D. A. Image method for efficiently simulation small-room acoustics // Journal of Acoustical Society of America. 1979. V. 64. ¹ 4. P. 943–950. 41. Cioffi J. M., Kailath T. Fast, recursive-least-squares transversal filters for adaptive filtering // IEEE Trans. Acoustics, Speech, and Signal Processing. 1984. V. 32. ¹ 2. P. 304–337. 42. Slock D. T. M., Kailath T. Numerically stable fast transversal filters for recursive least squares adaptive filtering // IEEE Trans. Signal Processing. 1991. V. 39. ¹ 1. P. 92–114. 43. Djigan V. I. Some tricks of calculations in MIL RLS algorithm // Proceedings of the 23-th International Conference on Digital Signal Processing and its Applications (DSPA-2021). Moscow, Russia, March 24 – 26, 2021. 4 p. 44. Djigan V. I. Low complexity RLS adaptive filters // Proceedings of the 23-th International Conference on Digital Signal Processing and its Applications (DSPA-2022). Moscow, Russia, March 30 – April 1, 2022. 5 p.
The article analyzes the main approaches to the visualization of GIS, highlights their advantages and disadvantages. A new approach is proposed and an appropriate algorithm for contrast visualization of small-sized, low-contrast objects is developed. The developed algorithm is based on calculating the contrast coefficient of the specified spectral characteristics of the object and background, finding local maxima of the spectral contrast function, selecting a set of spectral images from the corresponding maxima and synthesizing a grayscale or color image from this set. Examples and results of numerical studies confirming the effectiveness of the proposed approach are presented. 2. Pozhar V.E., Machikhin A.S., Gaponov M.I., Shirokov C.V., Mazur M.M., Sheryshev A.E. Hyperspectrometer based on tunable acousto-optic filters for UAVs // Lighting Engineering. 2018. No. 4. pp. 47-50. 3. Del Aguila A., Efremenko D.S., Trautmann T. Review of dimensionality reduction methods for processing hyperspectral optical signals // Lighting Engineering. 2019. No. 4. pp. 60-70. 4. Sheremetyeva T.A., Filippov G.N. Image conversion method // Russian Patent No. 2267232. 2005. 5. Know H., Der S. Z., Nasrabadi N. M. Adaptive multisensor target detection using feature-based fusion // Opt. Eng. 2002. Vol. 41, N 1. P. 69-80. 6. Borzov S.M. Potaturkin O.I. The choice of an informative system of signs in the classification of agricultural crops by hyperspectral data // Autometry. 2020. Vol. 56. No. 4. pp. 134-144. 7. Maltsev G.N., Kozinov I.A. Optimization of the number of spectral channels in the tasks of processing and analyzing hyperspectral data of remote sensing of the world ocean // Fundamental and applied hydrophysics. 2015. Vol. 8. No. 4. pp. 92-100. 8. Lozhkin L.D. Color, its measurement and perception in color television. Moscow: KomKniga, 2018. 480 p. 9. Cherepanov A.S. Vegetative indices // Geomatics. 2011. No. 2. pp. 98-102. 10. Savorsky V.P., Kashnitsky A.V., Konstantinova A.M., Balashov I.V., etc. Possibilities of analysis of hyperspectral indices in information systems of remote monitoring of the Constellation-Vega family // Modern problems of remote sensing of the Earth from space. 2016. Vol. 13. No. 3. pp. 28-45 11. Miroshnikov M.M. Theoretical foundations of optoelectronic devices. L.: Mechanical Engineering, 1977. 696 p.
The analysis of the possibilities of conventional processing methods for recognizing markings on images is carried out. In particular, the Fourier-Mellin transform, the Sift detector-descriptor together with the Ransac method, the contour extraction method, and the Hough transform were considered. A classification algorithm based on Erosion operation, threshold filtration and Hough transform is described. The need for an erosion operation, threshold filtration is due to the above-mentioned features of micro-labeling images. The classification used structural analysis of the Hough transform results, configured to search for straight line segments. As recognition criteria, it is proposed to use the proportion of registered segments with specified angles of mutual orientation or the local predominance of such segments in the histogram of the distribution. The efficiency of the proposed methods is established on different types of images of markings and background relief. Metrics for evaluating the effectiveness of the proposed solutions are given.
2. Gulyaev P. V., Shelkovnikov E. Yu. Search and recognition of the studied area in probe microscopy using markings applied by power lithography or nanoindentation // Chemical physics and mesoscopy. 2022. vol 24, No 1, pp. 111-120 3. Kislova TV. Investigation of the durability of nanomarkings applied to steel products using a scanning probe microscope under the influence of operational factors [In Russian]. NBI technologies 2021, vol. 15, no 1, pp. 29-33 4. Bessmeltsev V. P., Bulushev Å. D. Fast image registration algorithm for automated inspection of laser micromachining. Computer Optics, 2014, vol. 38, no. 2, pp. 343-350. [In Russian]. https://doi.org/10.18287/0134-2452-2014-38-2-343-350 5. Glumov N. I., Kuznetsov A. V. Copy-move image forensics detection. Computer Optics, 2011, vol. 35, no. 4, pp. 508-512. (In Russian). 6. Image Registration using Log-polar transformation, Phase correlation (Fourier-Mellin). URL: https://github.com/polakluk/fourier-mellin?ysclid=l71vcehrrb388493271 (Access date: 15.09.2022). 7. Krasnabayeu YA, Chistabayeu DV, Malyshev AL. Comparison of binary feature points descriptors of images under distortion conditions. [In Russian]. Computer Optics 2019; 43(3): 434-445. DOI: 10.18287/2412-6179-2019-43-3-434-445/. 8. Moisan L. Moulon P., Monasse P. Automatic Homographic Registration of a Pair of Images, with A Contrario Elimination of Outliers // Image Processing On Line.- 2012.- vol.2.- pp. 56–73. https://doi.org/10.5201/ipol.2012.mmm-oh 9. Moisan L. Moulon P., Monasse P. Automatic Homographic Registration of a Pair of Images, with A Contrario Elimination of Outliers // Image Processing On Line.- 2012.- vol.2.- pp. 56–73. https://doi.org/10.5201/ipol.2012.mmm-oh 10. DEMO. Automatic homographic registration of a pair of images, with a contrario elimination of outliers. URL: https://ipolcore.ipol.im/demo/clientApp/demo.html?id=60 (Access date: 15.09.2022). 11. Mironov V.L. The fundamentals of scanning probe microscopy // Institute of Physics of Microstructures. Nizhny Novgorod.- 2004.- 110s 12. Micromarks-recognition. URL: https://github.com/GulyaevPV/Micromarks-recognition (Access date: 15.09.2022).
V.A. Eremeev, e-mail: foton@rsreu.ru A.A. Makarenkov, e-mail: foton@rsreu.ru The Ryazan State Radio Engineering University (RSREU), Russia, Ryazan Keywords:
Proposed approach of using U-net with an addition of texture feature is assessed on the real images from Russian remote sensing system “Resurs-P”. Results of the performance evaluation showed that proposed approach produces from 5% to 14% increase (F-measure) of homogenous areas identification compared to the U-net network processing data without texture feature.
2. Haralick, R. M., Shanmugam, K. & Dinstein, I. Textural Features for Image Classification // IEEE Trans. Man, and Cybernetics 3, 1973. pp. 610-621. 3. Marina Sokolova, Guy Lapalme. A systematic analysis of performance measures for classification tasks // Information processing and management, 2009, vol. 45, no. 4, pp. 427-437.
The article reveals the algorithm for generating query metrics by content. Currently, there are metric and non-metric algorithms for assessing the similarity of images. Content evaluation was carried out using the MobileNet v2 convolutional neural network. The last layer of the neural network was used, convolution and subsampling operations were used, which allow you to convert images into a hash that is invariant to various distortions, which allows you to find identical images in the search database with high efficiency. The algorithm makes it possible to increase the probability of finding images based on their content by hashing each image from the search base using convolutional neural networks. The developed algorithm for detecting visual similarity of images does not use a priori information, but takes into account only image pixels. It can be used in solving problems of visual search, classification, as well as software for systems for searching attractions from photographs in travel companies, in text and image search systems, in open access electronic catalogs (local history museums, cadastral maps). The results of the algorithm presented in the article show its effectiveness. The comparison was carried out with known content search algorithms. The short running time and high efficiency of the presented algorithm make it possible to use such an application on a personal computer. To reduce the operating time of the algorithm for detecting moving objects, it is planned to use parallel computing technologies.
2. Ignatov A.K. Sposob indeksatsii i poiska tsifrovykh izobrazheniy // Patent na izobreteniye RU 2510935 C2, 10.04.2014. Zayavka ¹ 2011138862/08 ot 23.09.2011. 3. Rogov A. A., Rogova K. A., Spiridonov K.N., Bystrov M.YU. Informatsionno-poiskovaya sistema Petroglifov Karelii // Vestnik komp'yuternykh i informatsionnykh tekhnologiy. – 2008. – ¹ 6 (48). – S. 6-11. 4. Babenko A., Lempitskiy V. Effektivnyy algoritm poiska blizhayshikh sosedey pri bolshikh obyemakh poiskovoy bazy // Trudy 54-oy nauchnoy konferentsii MFTI. Innovatsii i vysokiye tekhnologii. Dolgoprudnyy: 2011. – S. 16-17. 5. El-saban M.A., Tavfik A.I., Chalabi A.A.M.T., Sayyed S.KH. Poisk izobrazheniy na yestestvennom yazyke // Patent na izobreteniye RU 2688271 C2, 21.05.2019. Zayavka ¹ 2016144699 ot 14.05.2015. 6. Noa Garcia, George Vogiatzis Learning Non-Metric Visual Similarity for Image Retrieval the IEEE International Conference on Computer Vision Workshops (ICCVW) – 2017 7. Venice Erin Liong, Jiwen Lu, Gang Wang, Pierre Moulin, and Jie Zhou. Deep hashing for compact binary codes learning. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition – 2015. 8. Albert Gordo, Jon Almazan, Jerome Revaud, and Diane Larlus. Deep image retrieval: Learning global representations for image search. In Proceedings of the IEEE European Conference on Computer Vision – 2016. 9. Wenjie Luo, Alexander G Schwing, and Raquel Urtasun. Efficient deep learning for stereo matching. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition – 2016. 10. Laurens van der Maaten Geoffrey Hinton Visualizing non-metric similarities in multiple maps. The Author(s) 2011. This article is published with open access at Springerlink.com DOI 10.1007/s10994-011-5273-4 11. Tomas Skopal On Fast Non-metric Similarity Search by Metric Access Methods. (Eds.): EDBT 2006, LNCS 3896, pp. 718–736 – 2006. 12. Potapova V., Tarasov A., Grinchenko N. Image Search by Content System Development Published in: 2018 IEEE East-West Design & Test Symposium (EWDTS) – 2018 – Ñ. 625-629. 13. Rudakov I.V., Vasyutovich I.M. Issledovaniye pertseptivnykh kheshfunktsiy izobrazheniya // V sb.: Nauka i obrazovaniye MGTU im. N.E. Baumana. S. 269–280. 14. Tzu-Heng Henry Lee. Wavelet Analysis for Image Processing. http://disp.ee.ntu.edu.tw/henry/wavelet_analysis.pdf 15. Nikolenko S., Kadurin A., Arkhangel'skaya Ye. Glubokoye obucheniye pogruzheniye v mir neyronnykh setey. – SPb.: Piter, 2019. – 480 c. 16. Tarasova V. YU., Tarasov A. S., Grinchenko N. N. Algoritm obnaruzheniya sovpadeniy v kollektsii izobrazheniy // v sbornike trudov Mezhdunarodnogo nauchno-tekhnicheskogo foruma STNO-2019. – Tom 4. – S. 164-168. 17. Bodrov O.A., Tarasov A.S., Tarasova V.Y., Bodrova I.V. IMAGE SEARCH ALGORITHM IN LOCAL DATA BASE 8th Mediterranean Conference on Embedded Computing, MECO 2019 - Proceedings. – 2019. – Ñ. 453-455. 18. Eric J. Stollnitz, Tony D. DeRose, David H. Salesin Wavelets for Computer Graphics Theory and Applications 19. F. Sabahi; M. Omair Ahmad;MNS Swamy Content-based Image Retrieval using Perceptual Image Hashing and Hopfield Neural Network. 2018 IEEE 61st International Midwest Symposium on Circuits and Systems (MWSCAS)
The formation of the cutting band is based on the placement of mutually displaced zeros of the system function on the unit circle. The formation of the bandwidth is based on the methods of linear programming from the condition of minimizing the transition bandwidth of the filter at a given value of the maximum error in the bandwidth. As a result, the proposed method for the synthesis of digital non-recursive rejection filters allows for a given filter range to obtain the specified parameters of passband and notch at the minimum possible width of the transition band. A comparative analysis of the quality of interference rejection by filters synthesized by the proposed and known methods according to the criteria of the interference transmission coefficient and the coefficient of subsurface visibility is carried out. The comparative analysis confirmed the possibilities of the proposed method to synthesize effective rejection filters with high indicators of the quality of Doppler signal isolation against the background of correlated interference. The simplest synthesis method based on the Fourier series expansion with the subsequent introduction of a weight function, due to its simplicity, can be used in the absence of strict requirements for filter parameters, while providing acceptable results. 2. Richards M.A., Scheer J.A., Holm W.A. (Eds.). Principles of Modern Radar: Basic Principles. New York: SciTech Publishing, IET, Edison. 2010. – 924 p. 3. Melvin W. L., Scheer J.A. (Eds.). Principles of Modern Radar: Advanced Techniques. New York: SciTech Publishing, IET, Edison, 2013. – 846 p. 4. Radar Handbook / Ed. by M.I. Skolnik. 3rd ed. McGraw–Hill, 2008. 1352 p. 5. Popov D.I. Adaptacija nerekursivnyh rezhektornyh fil'trov // Izvestija vuzov. Ra-diojelektronika. 2009. vol. 52. no. 4. P. 46-55. (in Russian). 6. Popov D.I. Autocompensation of the Doppler phase of clutter // Cifrovaja obrabotka signalov. 2009. no 2. pp. 30–33. (in Russian). 7. Popov D.I. Adaptive suppression of clutter // Cifrovaja obrabotka signalov. 2014. no. 4. pp. 32-37. (in Russian). 8. Popov D.I. Adaptivnije regektornjie filtrij kaskadnogo tipa // Cifrovaya obrabotka signalov. 2016. no. 2. pp. 53-56. (in Russian). 9. Popov D.I. Adaptive notch filter with real weights // Cifrovaya obrabotka signalov. 2017. no. 1. pp. 22-26. (in Russian). 10. Popov D.I. Optimizacja nerekursivnjih regektornjie filtrov s chastichnoj adaptaciej // Cifrovaya obrabotka signalov. 2018. no. 1. pp. 28-32. (in Russian). 11. Popov D.I. Optimizacija rezhektornyh fil'trov po verojatnostnomu kriteriju // Cifrovaja obrabotka signalov. 2021. no. 1. P. 55-58. (in Russian). 12. Rabiner L., Gould B. Teorija i primenenie cifrovoj obrabotki signalov (Theory and application of digital signal processing). – M.: Mir, 1978. – 848 p. (in Russian).
2. Shannon C. Probability of error for optimal codes in Gaussian channel. //Bell System Techn. J., May, 1959, Ð. 611-656. 3. Kotelnikov V.A. The Theory of Optimum Noise Immunity. McGraw-Hill Book Co., NY, 1959, p. 140. 4. Bykhovskiy M.A. Giperfazovaya modulyatsiya – optimal'nyy metod peredachi soobshcheniy v gaussovskom kanale svyazi. (Hyperphase Modulation - the optimal method of message transmission in the Gaussian communication channels). Moscow, Technospera publishers, 2018, Ð. 310. 5. G. A. Korn, T. M. Korn, Mathematical Handbook for Scientists and Engineers. New York/San Francisco,1968, McGraw-Hill Book Company, Ð. 1151. 6. Fikhtengolts G.M. Kurs differentsial'nogo i integral'nogo ischisleniya. (Course of differential and integral calculus). vol. 2. Moscow: Nauka, 1969, P. 806. 7. Proakis John G., Salehi M. Digital Communications (5th ed), McGraw-Hill Book Higher Education, 2008, Ð. 1150. 8. Error-Correcting Codes. Error-Correcting Codes, Second Edition. W. Wesley Peterson & E. J. Weldon. MIT Press, 1972, P. 560. 9. Vargauzin V.A., Tsikin I.A. Metody povysheniya energeticheskoy i spektral'noy effektivnosti tsifrovoy svyazi. (Vargauzin V.A., Tsikin I.A. Methods for improving the energy and spectral efficiency of digital radio communications). St. Petersburg: BHB-Petersburg, 2013, Ð. 352. 10. Ungerboeck G. Channel coding with multilevel phase signals. //IEEE Trans. Inf. Theory, 1982, vol. 28, no. 1, Ð. 55-67. 11. Caire G., Tarricco G., Biglieri E. Bit-Interleaved Coded Modulation. //IEEE Trans. Inf. Theory, 1998, vol. 44, no. 5, Ð. 927-946. 12. Wachmann U., Fischer R.F., Huber J.B. Multilevel codes: Theoretical concepts and practical design rules. //IEEE Trans. Inf. Theory, 1999, vol. 45, no. 5, Ð. 1361-1391.
It is shown that three different types of functions are used to process various parameters: the logistic sigmoid is used to predict the positioning accuracy and the bounding boxes vertical and horizontal offsets, the exponent is used to predict the height and width of boxes, and the SoftMax function is used to predict class probabilities. Formulas for the derivatives of these functions, which are required when training the network, are given. The multiple execution of given transformations, required for the YOLO detector operation, needs the calculation of an exponential function values. Methods of calculus mathematics make it possible to obtain a fairly accurate approximation of the exponential for a certain number of terms of the corresponding series taken into account. Nevertheless, to ensure the necessary accuracy of the correspondence between these transformations and operations performed on hardware, an unacceptably high computing power is required. We propose a different approach that does not require an exact approximation of given transformations. The logistic function is replaced by a rational sigmoid, the exponent is replaced by a shifted fourth-order parabolic curve. The modified SoftMax function also uses a shifted parabola. For the proposed transformations, formulas for calculating the loss back-propagation are analytically derived. Using the obtained formulas, a YOLO detector based on a modified neural network was trained. The modified YOLO detector with the proposed functions is implemented on a specially designed Xilinx FPGA board. Experimental studies of the board showed the high speed of the detector (more than 60 frames per second) and the high quality of object detection and classification. When processing typical video recordings, the values of the output array of floating-point numbers obtained on the video card of a personal computer coincided with the corresponding values calculated by hardware on the board with an error of less than 1%. The effectiveness of the proposed transformations is confirmed by the qualitative characteristics of object recognition, which are not inferior to the characteristics of the original detector. 2. Redmon J., Divvala S., Girshick R., Farhadi A. You Only Look Once: Unified, Real-Time Object Detection // Proceedings of the IEEE conference on computer vision and pattern recognition. 2016. pp. 779-788. 3. Redmon J., Farhadi A. YOLO9000: Better, Faster, Stronger // Proceedings of the IEEE conference on computer vision and pattern recognition. 2017. pp. 7263-7271. 4. Redmon J., Farhadi A. YOLOv3: An Incremental Improvement [Web-site] // arXiv. URL: https://arxiv.org/pdf/1804.02767.pdf (22.07.2020). 5. Rezatofighi H. et al. Generalized intersection over union: A metric and a loss for bounding box regression // Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. – 2019. – pp. 658-666. 6. Shalev-Shwartz S., Ben-David S. Understanding machine learning: From theory to algorithms. – Cambridge university press. – 2014. – 449 pp. 7. Backpropagation with Cross-Entropy and Softmax [Web-site] // ML Dawn. 2021. URL: https://www.mldawn.com/back-propagation-with-cross-entropy-and-softmax/ (20.01.2022). 8. The Derivative of Softmax Function [Web-site] // ML Dawn. 2021. URL: https://www.mldawn.com/the-derivative-of-softmaxz-function-w-r-t-z/ (20.01.2022). 9. Buckland M., Gey F. The relationship between recall and precision // Journal of the American society for information science. – 1994. – Vol. 45. – No. 1. – pp. 12-19. 10. Van Trees H. L. Detection, estimation, and modulation theory, part I: detection, estimation, and linear modulation theory. – John Wiley & Sons, 2004, 716 pp. |