PARALLEL FPGA-BASED PROCESSING FOR IMPROVED INFRARED TARGET DETECTION USING VMD AND SPWVD

Infrared sensors are widely used in target detection due to their small size and low cost. But the signal-to-noise ratio (SNR) is very low and the range of action is short. In order to increase the probability of detecting a specific target, noise must be more thoroughly suppressed and signal features must be extracted more accurately. Algorithms such as variational mode decomposition (VMD) and smoothed pseudo-Wegener distribution (SPWVD) that have emerged in recent years can be used to solve this problem. VMD works well in suppressing noise in non-stationary and non-linear signals. SPWVD has high time-frequency resolution and clean spectra. However, infrared detection requires high real-time performance, so these methods are limited in use due to the large amount of calculation. This article builds a highly parallel processing system based on FPGA and introduces VMD and SPWVD into the scene of real-time infrared detection of moving targets. It meets real-time requirements while greatly improving detection accuracy.