Implementation of Spatial Image Filtering Function on a Vector Coprocessor
Abstract
To increase the performance of universal microprocessors of MIPS-like architecture, NIISI RAS developed a specialized coprocessor that allows accelerating operations with complex and real numbers of single and double precision. The article presents the results of using a 128-bit vector coprocessor for the image filtering task. Using the example of two vectorization options, the increase in the efficiency of calculations when solving this problem is show
About the Authors
S. I. AryashevRussian Federation
P. A. Chibisov
Russian Federation
V. V. Tsvetkov
Russian Federation
D. A. Trubitsyn
Russian Federation
K. A. Petrov
Russian Federation
References
1. П.Б. Богданов, О.Ю. Сударева. Применение отечественных специализированных процессоров семейства КОМДИВ в научных расчетах // Информационные технологии и вычислительные системы 3/2016.
2. S. Sadangi, S. Baraha, D. K. Satpathy and P. K. Biswal, "FPGA implementation of spatial filtering techniques for 2D images," 2017 2nd IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT), Bangalore, India, 2017, pp. 1213-1217, doi: 10.1109/RTEICT.2017.8256791.
3. D. H. Rao and P. P. Panduranga, "A Survey on Image Enhancement Techniques: Classical Spatial Filter, Neural Network, Cellular Neural Network, and Fuzzy Filter," 2006 IEEE International Conference on Industrial Technology, Mumbai, India, 2006, pp. 2821-2826, doi: 10.1109/ICIT.2006.372671.
4. Н.С. Сергеев. Пространственная фильтрация изображений в системах технического зрения.
5. П.Б. Богданов, О.Ю. Сударева. Декодирование изображений в формате JPEG на процессорах КОМДИВ // Информационные технологии и вычислительные системы 1/2019.
Review
For citations:
Aryashev S.I., Chibisov P.A., Tsvetkov V.V., Trubitsyn D.A., Petrov K.A. Implementation of Spatial Image Filtering Function on a Vector Coprocessor. SRISA Proceedings. 2023;13(4):85-90. (In Russ.)