In this work, we present a FPGA-based Generalized Hough Transform custom processor to calculate similarities between arbitrary shapes. Raw data are 44x36 DC images extracted directly from low-resolution compressed video (352x288). The outputs are two numbers per frame that quantify the image similitude in terms of scale and rotation. The proposed architecture efficiently resolves the detection of pixel pairs, and the voting of distances and rotations, without memory access conflicts. These operations are inherent to Hough transformation. The paper condenses some circuit solutions suitable to hardwiring video processing. They take full advantage of using small embedded memories as look-up tables. The complete processor is validated with benchmark video samples that cover different scenarios and problems: sport, drama, and news. The final version internally operates at 100 MHz and fits inside a small FPGA chip. The highly concurrent architecture employs both pipelining and parallelism using hardware replication. The final performance is over 40 Giga fixed-point operations per second.
1 January 2024 in IEEE Transactions on Circuits and Systems I: Regular Papers
This study presents a Flexible Test Bench (FTB) implemented with FPGA that synthesises of a large number of strategies that apply “Spread Spectrum” to reduce the energy of the fundamental harmonics present in a conventional Pulse Width Modulation (PWM). The FTB not only incorporates most of the known spread spectrum techniques but also allows to combine them and even to easily create new ones, thus providing a highly flexible test bench. The FTB has been described in standard VHDL, so it can be synthesised using any synthesis tool, and the number of logical resources used by each of them can be determined according to the spread spectrum strategy configured in its registers. Therefore, a tool is available to choose, according to the spectral response and the consumption of logic resources, the most suitable PWM spread spectrum strategy for each application. The particularity of the proposal is to have a unique hardware platform that allows the comparison of different techniques under the same physical implementation conditions since geometry and physical location are important when evaluating the conducted and radiated emission of the circuits. Another objective of our FTB is to solve in a unified way all the numerical aspects present in the FPGA-based implementation so that the numerical ranges and roundings used affect all the implemented techniques equally.