Implementing large-kernel 2-D filters using Impulse CoDeveloper

Bidimensional convolution is a low-level processing algorithm which is of great interest in many areas, but its high computational cost limits the size of the kernels, especially in real-time embedded systems. This work describes the process of designing 2-D filters with large kernels using the Impulse CoDeveloper™ electronic system-level tool by Impulse Accelerated Technologies. The proposed design includes an efficient management of the operations at the borders of the input array. Several kernel sizes have been tested, ranging from 20×20 coefficients to 50×50 coefficients, with different bits-per-pixel configurations for both the kernel coefficients and the input data. In each case, performance is reported in terms of area utilization and minimum clock period.