Designing a Classifier for Active Fire Detection from Multispectral Satellite Imagery Using Neural Architecture Search

TL;DR

This paper employs reinforcement learning-based Neural Architecture Search to design a compact neural network for active fire detection from multispectral satellite imagery, optimized for low-power nanosatellite onboard processing.

Contribution

It introduces a novel NAS approach that uses a regression-based reward function to efficiently design resource-constrained neural networks for satellite fire detection.

Findings

Designed a neural network with 1,716 parameters

Achieved inference latency of 984 microseconds

Power consumption around 800mW during inference

Abstract

This paper showcases the use of a reinforcement learning-based Neural Architecture Search (NAS) agent to design a small neural network to perform active fire detection on multispectral satellite imagery. Specifically, we aim to design a neural network that can determine if a single multispectral pixel is a part of a fire, and do so within the constraints of a Low Earth Orbit (LEO) nanosatellite with a limited power budget, to facilitate on-board processing of sensor data. In order to use reinforcement learning, a reward function is needed. We supply this reward function in the shape of a regression model that predicts the F1 score obtained by a particular architecture, following quantization to INT8 precision, from purely architectural features. This model is trained by collecting a random sample of neural network architectures, training these architectures, and collecting their classification performance statistics. Besides the F1 score, we also include the total number of trainable parameters in our reward function to limit the size of the designed model and ensure it fits within the resource constraints imposed by nanosatellite platforms. Finally, we deployed the best neural network to the Google Coral Micro Dev Board and evaluated its inference latency and power consumption. This neural network consists of 1,716 trainable parameters, takes on average 984{\mu}s to inference, and consumes around 800mW to perform inference. These results show that our reinforcement learning-based NAS approach can be successfully applied to novel problems not tackled before.