# Ferroelectric Optoelectronic Sensor for Intelligent Flame Detection and In-Sensor Motion Perception

## TL;DR

A new sensor array detects weak UV signals from flames and uses neural networks to accurately recognize flame motion and signals.

## Contribution

A ferroelectric optoelectronic sensor array with multimode functionality for precise UV detection and motion recognition in flame safety systems.

## Key findings

- The sensor array detects ultraweak UV signals with high accuracy through ferroelectric modulation.
- A lightweight convolutional neural network achieves 96.47% accuracy in flame motion recognition.
- An optoelectronic artificial neural system identifies flame optical signals with 90.51% accuracy.

## Abstract

The Ga₂O₃/In₂Se₃ heterojunction ferroelectric optoelectronic sensor array enables precise detection of ultraweak UV signals through ferroelectric modulation.Efficient flame detection across all time periods is achieved through terminal devices and cloud-based alert systems.The lightweight convolutional neural network-based approach achieves a flame motion recognition accuracy of 96.47%, while the optoelectronic artificial neural system attains 90.51% accuracy in identifying flame optical signals.

The Ga₂O₃/In₂Se₃ heterojunction ferroelectric optoelectronic sensor array enables precise detection of ultraweak UV signals through ferroelectric modulation.

Efficient flame detection across all time periods is achieved through terminal devices and cloud-based alert systems.

The lightweight convolutional neural network-based approach achieves a flame motion recognition accuracy of 96.47%, while the optoelectronic artificial neural system attains 90.51% accuracy in identifying flame optical signals.

The online version contains supplementary material available at 10.1007/s40820-025-01968-x.

Next-generation fire safety systems demand precise detection and motion recognition of flames. In-sensor computing, which integrates sensing, memory, and processing capabilities, has emerged as a key technology in flame detection. However, the implementation of hardware-level functional demonstrations based on artificial vision systems in the solar-blind ultraviolet (UV) band (200–280 nm) is hindered by the weak detection capability. Here, we propose Ga2O3/In2Se3 heterojunctions for the ferroelectric (abbreviation: Fe) optoelectronic sensor (abbreviation: OES) array (5 × 5 pixels), which is capable of ultraweak UV light detection with an ultrahigh detectivity through ferroelectric regulation and features in configurable multimode functionality. The Fe-OES array can directly sense different flame motions and simulate the non-spiking gradient neurons of insect visual system. Moreover, the flame signal can be effectively amplified in combination with leaky integration-and-fire neuron hardware. Using this Fe-OES system and neuromorphic hardware, we successfully demonstrate three flame processing tasks: achieving efficient flame detection across all time periods with terminal and cloud-based alarms; flame motion recognition with a lightweight convolutional neural network achieving 96.47% accuracy; and flame light recognition with 90.51% accuracy by means of a photosensitive artificial neural system. This work provides effective tools and approaches for addressing a variety of complex flame detection tasks.

The online version contains supplementary material available at 10.1007/s40820-025-01968-x.