Abstract
In this work, a computationally efficient implementation of the Brady algorithm for fast Hough transform (FHT) is built for the Russian coprocessor CPCA, which is a part of the 1890VM9Ya “KOMDIV128-M” system-on-chip. It is shown that FHT is widely used in image analysis, from vision systems of unmanned vehicles to computed X-ray tomography. The classical recursive implementation of FHT is analyzed from a low-level perspective. For the first time, a more efficient non-recursive version of the algorithm is considered. The workload on arithmetic logic units and coprocessor memory is analyzed and experimental performance measurements are carried out. It is shown that the theoretically possible performance of the non-recursive algorithm on CPCA is 800 Mops. The maximum performance achievable in practice is 470 Mops and the maximum experimental value is 406 Mops. At the same time, the workload on the coprocessor units reaches 18%. Thus, despite the relatively small number of arithmetic operations in the method, the use of the coprocessor proves reasonable.
Similar content being viewed by others
REFERENCES
Brady, M.L., A fast discrete approximation algorithm for the radon transform, SIAM J. Comput., 1998, vol. 27, no. 1, pp. 91–99.
Nikolaev, D.P., Karpenko, S.M., and Nikolayev, I.P., Hough transform: Underestimated tool in the computer vision field, Proc. 22nd Eur. Conf. Modelling and Simulation, 2008, pp. 238–243.
Hough, P.V.C., Machine analysis of bubble chamber pictures, Conf. Proc. C, 1959, vol. 590914, pp. 554–558.
Duda, R.O. and Hart, P.E., Use of the Hough transformation to detect lines and curves in pictures, Commun. ACM, 1972, vol. 15, no. 1, pp. 11–15.
Asvatov, E.N., Ershov, E.I., and Nikolaev, D.I., Robust orthogonal linear regression for low-dimensional histograms, Sens. Sist., 2017, vol. 31, no. 4, pp. 331–342.
Nikolaev, D.P. and Nikolayev, P.P., Linear color segmentation and its implementation, Comput. Vision Image Understanding, 2004, vol. 94, no. 1, pp. 115–139.
Kunina, I.A., Gladilin, S.A., and Nikolaev, D.P., Blind compensation of radial distortion in a single image using fast Hough transform, Comput. Opt., 2016, vol. 40, no. 3, pp. 395–403.
Nur Shazwani Aminuddin, Masrullizam Mat Ibrahim, Nursabillilah Mohd Ali, et al., A new approach to highway lane detection by using Hough transform technique, J. Inf. Commun. Technol., 2017, vol. 16, no. 2, pp. 244–260.
Kunina, I.A., Panfilova, E.I., and Povolotskii, M., Detection of pedestrian crossings on road images based on a dynamic alignment method for time series, Tr. Inst. Sist. Program. Ross. Akad. Nauk (Proc. Inst. Syst. Program. Russ. Acad. Sci.), 2018, vol. 68, no. S1, pp. 23–31.
Panfilova, E.I., Shipitko, O.S., and Kunina, I.A., Fast Hough transform-based road markings detection for autonomous vehicle, Proc. 13th Int. Conf. Machine Vision (ICMV), 2021, vol. 11605, pp. 671–680.
Jahan, R., Suman, P., and Singh, D.K., Lane detection using canny edge detection and Hough transform on raspberry Pi, Int. J. Adv. Res. Comput. Sci., 2018, vol. 9, no. S2, pp. 85–89.
Guan, J., An, F., Zhang, X., et al., Energy-efficient hardware implementation of road-lane detection based on Hough transform with parallelized voting procedure and local maximum algorithm, IEICE Trans. Inf. Syst., 2019, vol. E102.D, no. 6, pp. 1171–1182.
Thongpan Narathip, Rattanasiriwongwut Montean, and Ketcham Mahasak, Lane detection using embedded system, Int. J. Comput., Internet Manage., 2020, vol. 28, no. 2, pp. 46–51.
Kotov, A.A., Konovalenko, I.A., and Nikolaev, D.P., Object tracking in a video stream, optimized using fast Hough transform, Inf. Tekhnol. Vychisl. Sist., 2015, no. 1, pp. 56–68.
Bocharov, A.D., Linear regression method tolerant to extreme stationary noise, Sens. Sist., 2020, vol. 34, no. 1, pp. 44–56.
Green, A.I., Schwartz, R., Dodge, J., Smith, N.A., and Etzioni, O., Commun. ACM, 2020, vol. 63, no. 12, pp. 54–63.
Tropin, D.V., Ilyuhin, S.A., Nikolaev, D.P., and Arlazarov, V.V., Approach for document detection by contours and contrasts, 2020, arXiv:2008.02615 [cs.CV].
Bezmaternykh, P.V. and Nikolaev, D.P., A document skew detection method using fast Hough transform, Proc. 12th Int. Conf. Machine Vision (ICMV), 2020, vol. 11433, pp. 132–137.
Gaier, A.V. and Sheshkus, A.V., Neural network-based detection of the top and base lines of a text line in an image, XII Mul’tikonferentsiya po problemam upravle-niya (Proc. 13th Multiconf. Management Problems), 2019, pp. 53–58.
Limonova, E., Bezmaternykh, P., Nikolaev, D., and Arlazarov, V., Slant rectification in Russian passport OCR system using fast Hough transform, Proc. 9th Int. Conf. Machine Vision (ICMV), 2017, vol. 10341, pp. 127–131.
Martynov, S.I. and Bezmaternykh, P.V., Aztec core symbol detection method based on connected components extraction and contour signature analysis, Proc. 12th Int. Conf. Machine Vision (ICMV), 2020, vol. 11433, pp. 27–34.
Bulatov, K.B., Chukalina, M.V., and Nikolaev, D.P., Fast X-ray sum calculation algorithm for computed tomography, Bull. South Ural State Univ., Ser.: Math. Modell. Program. Comput. Software, 2020, vol. 13, no. 1, pp. 95–106.
Dolmatova, A.V., Chukalina, M.V., and Nikolaev, D.P., Accelerated FBP for computed tomography image reconstruction, Proc. IEEE Int. Conf. Image Processing, 2020, pp. 3030–3034.
Ingacheva, A.S., Sheshkus, A.V., Chernov, T.S., et al., X-ray computed tomography: A new tool in recognition, Tr. Inst. Sist. Program. Ross. Akad. Nauk (Proc. Inst. Syst. Program. Russ. Acad. Sci.), 2018, vol. 68, no. S1, pp. 90–99.
Bulatov, K., Chukalina, M., Buzmakov, A., et al, Monitored reconstruction: Computed tomography as an anytime algorithm, IEEE Access, 2020, vol. 8, pp. 110759–110774.
Sheshkus, A., Chirvonaya, A., Nikolaev, D., and Arlazarov, V.L., Vanishing point detection with direct and transposed fast Hough transform inside the neural network, Comput. Opt., 2020, vol. 44, no. 5, pp. 737–745.
Sheshkus, A. and Nikolaev, D., Houghencoder: Neural network architecture for document image semantic segmentation, Proc. IEEE Int. Conf. Image Processing, 2020, pp. 1946–1950.
Lin, Y., Pintea, S.L., and van Gemert, J.C., Deep Hough-transform line priors, Comput. Vision, 2020, pp. 323–340.
Han, Q., Zhao, K., Xu, J., and Cheng, M.-M., Deep Hough transform for semantic line detection, Comput. Vision, 2020, pp. 249–265.
Teplyakov, L., Kaymakov, K., Shvets, E., and Nikolaev, D., Line detection via a lightweight CNN with a Hough layer, Proc. 13th Int. Conf. Machine Vision (ICMV), 2021, vol. 11605, pp. 376–385.
Usilin, S.A., Arlazarov, V.V., Putintsev, D.N., and Tarkhanov, I.A., Methods for image recognition and processing during the construction of oil and gas wells, Inf. Tekhnol. Vychisl. Sist., 2020, no. 1, pp. 12–24.
Zhao, H. and Zhang, Z., Improving neural network detection accuracy of electric power bushings in infrared images by Hough transform, Sensors, 2020, vol. 20, no. 10, pp. 1–16.
Ershov, E.I. and Karpenko, S.M., Fast Hough transform and approximation properties of dyadic patterns, 2017.
Ershov, E.I., Terekhin, A.P., and Nikolaev, D.P., Generalization of the fast Hough transform for three-dimensional images, Inf. Protsessy, 2017, vol. 17, no. 4, pp. 294–308.
Mehta, D.P. and Sahni, S., Handbook of Data Structures and Applications, Chapman & Hall/CRC, 2018, 2nd ed.
Mikrosistema integral’naya 1890VM9Ya. Ukazaniya po primeneniyu (Integrated Microsystem 1890VM9Ya: Application Notes), Moscow: Sci. Res. Inst. Syst. Anal. Russ. Acad. Sci., 2016.
Sudareva, O.Yu., Effektivnaya realizatsiya algoritmov bystrogo preobrazovaniya Fur’e i svertki na mikroprotsessore KOMDIV128-RIO (Efficient Implementation of Fast Fourier Transform and Convolution Algorithms on the KOMDIV128-RIO Microprocessor), Moscow: Sci. Res. Inst. Syst. Anal. Russ. Acad. Sci., 2014.
Raiko, G.O., Mel’kanovich, V.S., and Pavlovskii, Yu.A., Programming technology for multiprocessor processing of hydroacoustic signals on computing devices of the KOMDIV family, Gidroakust., 2014, vol. 2, no. 20, pp. 85–92.
Funding
This work was supported in part by the Russian Foundation for Basic Research, project nos. 18-29-26017 and 18-29-26027.
Author information
Authors and Affiliations
Corresponding authors
Additional information
Translated by Yu. Kornienko
Rights and permissions
About this article
Cite this article
Anikeev, F.A., Raiko, G.O., Limonova, E.E. et al. Efficient Implementation of Fast Hough Transform Using CPCA Coprocessor. Program Comput Soft 47, 335–343 (2021). https://doi.org/10.1134/S0361768821050029
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S0361768821050029