UAV Detection: A STDP trained Deep Convolutional Spiking Neural Network Retina-Neuromorphic Approach

TL;DR

This paper introduces a novel UAV detection method using a dynamic vision sensor and a spiking neural network trained with STDP, emphasizing the potential of neuromorphic vision for real-time drone identification.

Contribution

It is the first work to utilize an event camera combined with an STDP-trained deep convolutional SNN for UAV detection, highlighting its advantages and future prospects.

Findings

Effective UAV detection using event-based vision data.

Comparison of real and simulated data to evaluate approach.

Discussion of current limitations and future device improvements.

Abstract

The Dynamic Vision Sensor (DVS) has many attributes that allow it to be well suited to the task for UAV Detection. This paper is the first to look at exploiting the features of an Event Camera solely for Drone Detection while combining it with a Spiking Neural Network (SNN) trained using the unsupervised approach of Spike-Time-Dependent Plasticity (STDP) for feature and pattern recognition for detection. Highlighting the key features and current drawbacks with the technology while comparing real and simulated data to show how future devices could overcome these drawbacks to help tackle this current problem.