RTX's Raytheon demonstrates event-based infrared camera for real-time tracking

Defense, surveillance, and aerospace sectors require sensors capable of detecting and tracking fast-moving objects under complex conditions. Raytheon, an RTX business, has demonstrated an event-based mid-wave infrared (MWIR) camera designed to address these challenges by fundamentally changing how infrared data is captured and processed.

Unlike conventional infrared cameras that record full image frames at fixed intervals, the new system operates using an event-based architecture. It detects and transmits only pixel-level changes in motion, generating a continuous stream of events rather than complete images. This reduces the volume of data generated and lowers processing and power requirements, while maintaining high temporal resolution.

During a field demonstration in Northern California, the camera successfully tracked multiple dynamic targets, including ground vehicles, aircraft, and active fire sources. By focusing exclusively on motion, the system captured rapid changes that are often missed or delayed in traditional frame-based infrared imaging, where processing occurs after full-frame acquisition.

Event-based sensing for faster detection
The key innovation lies in the sensor’s ability to respond asynchronously to changes in the scene. Each pixel operates independently, triggering only when it detects a variation in infrared intensity. This approach eliminates redundant data from static areas of the image and enables near-instantaneous detection of movement.

From an engineering perspective, this architecture significantly reduces latency between detection and output. Lower data throughput also decreases the computational load on downstream processors, which can be critical in systems with limited onboard processing capacity, such as unmanned aerial vehicles or edge-based defense platforms.

Performance advantages in complex environments
Event-based MWIR sensing is particularly suited to environments characterized by high-speed motion, visual clutter, or rapidly changing thermal signatures. Traditional systems may struggle in such conditions due to frame rate limitations or the need to process large volumes of data before identifying relevant changes.

By contrast, the demonstrated system provides continuous updates on motion events, enabling more responsive tracking and potentially improving reaction times in applications such as threat detection or target interception. The ability to detect subtle or rapid thermal variations also enhances situational awareness in low-visibility or high-noise environments.

Applications in defense and security systems
The technology is relevant for a range of defense and national security applications where speed and data efficiency are critical. These include missile guidance systems requiring rapid target acquisition, airborne and space-based surveillance platforms, and ground-based monitoring systems for perimeter and base protection.

In addition, the reduced power and processing demands make the sensor architecture suitable for integration into compact or distributed systems, where energy efficiency and real-time performance are key design constraints.

Development under neuromorphic sensing program
The camera was developed as part of the Defense Advanced Research Projects Agency’s Fast Event-based Neuromorphic Camera and Electronics (FENCE) program. This initiative focuses on advancing neuromorphic sensing technologies that mimic biological vision systems by prioritizing dynamic changes over static information.

With the initial development phase completed, further demonstrations are expected to evaluate the sensor across a broader range of operational scenarios and target conditions, supporting potential integration into next-generation sensing platforms.

Edited by Natania Lyngdoh, Induportals Editor — Adapted by AI.

www.rtx.com