Prism™ A-ISR is AI-enabled airborne intelligence, surveillance, and reconnaissance (ISR) software for real-time ground object detection, classification, and tracking from uncrewed aircraft systems (UAS) and other aerial platforms. Using thermal infrared and electro-optical (EO) sensor data, Prism A-ISR supports air-to-ground object detection, AI-based object classification, and real-time tracking in cluttered or degraded visual environments. It combines patented image processing, AI perception models, multispectral training data, and edge processing expertise to improve long-range situational awareness while reducing false alarms and integration risk.
The software supports passive, multi-object detection and tracking for persistent surveillance, force protection, and intelligence missions. Custom object classes and updated AI models can be generated with AIMMGen™ synthetic model-training services, helping developers adapt to evolving operational needs.
Built for OEM integration and scalable deployment, Prism A-ISR supports 14-bit longwave infrared (LWIR) and midwave infrared (MWIR) data from Teledyne FLIR Boson® and Neutrino® thermal camera families, including fixed focal length and continuous zoom (CZ) configurations, as well as third-party infrared and visible EO cameras. Designed for NVIDIA Orin and Qualcomm 8550 embedded AI platforms, Prism A-ISR helps developers add mission-ready EO/IR perception and real-time object tracking to next-generation airborne ISR systems.
The software supports passive, multi-object detection and tracking for persistent surveillance, force protection, and intelligence missions. Custom object classes and updated AI models can be generated with AIMMGen™ synthetic model-training services, helping developers adapt to evolving operational needs.
Built for OEM integration and scalable deployment, Prism A-ISR supports 14-bit longwave infrared (LWIR) and midwave infrared (MWIR) data from Teledyne FLIR Boson® and Neutrino® thermal camera families, including fixed focal length and continuous zoom (CZ) configurations, as well as third-party infrared and visible EO cameras. Designed for NVIDIA Orin and Qualcomm 8550 embedded AI platforms, Prism A-ISR helps developers add mission-ready EO/IR perception and real-time object tracking to next-generation airborne ISR systems.
ENHANCED DETECTION AND TRACKING
Extend detection range, improve track persistence, and reduce false alarms
- Patented denoise and super-resolution improves signal to noise
- Electronic stabilization and dehaze algorithms improve image quality
- Patented denoise algorithm minimizes bit rate
STREAMLINE LWIR, MWIR, AND EO CAMERA INTEGRATION
Native support for Boson+, Boson SX8, and Neutrino series accelerates deployment
- Supports 640 × 512 and 1280 × 1024 LWIR and MWIR resolutions
- Compatible with fixed focal length and CZ lens configurations
- Accepts 14-bit IR-camera data
- Delivers frame-rate perception on EO resolutions up to 4 MP
BUILT FOR DEVELOPERS AND DEPLOYMENT
Lower integration risk with production-ready technology, developer tools, and expert technical support
- Sample C++ reference application, libraries, and documentation
- Experienced technical services team supports integration and deployment
- Operates on Qualcomm and NVIDIA embedded hardware
- Adapt models for new missions with Prism AIMMGen™ synthetic data
Specifications
Performance Metric
- Included Filters
- Denoise, Super Resolution, Electronic Stabilization
- Inference Time
- ~10 ms
- Input Wavelength
- EO, MWIR, and LWIR
- Metadata Output
- Bounding Box Area
Bound Box Location
Category ID
Object Class Label
Confidence
Time Stamp
Track ID
Frame Number
- Object Classes - Infrared and EO
- Person, vehicle (umbrella class car, bus, truck, van, construction), boat (umbrella class for all types of vessels on water)
- Object Classification >60% Confidence
- ≥10x10 Pixels on Target (minimum)
- Simultaneous Targets Tracked
- Maxuimum is System Dependent, Default is 16, Configurable
- Target Detection Confidence Threshold
- ≥60% for Small Targets (<32 x 32 pixels), Configurable
System Requirement
- Deliverables
- • User Graphical Interface (GUI)
• Software Development Kit (SDK) with Prism A-ISR solution package
• Technical Documentation
- EO Camera
- 4 MP with UVC Input Typical
(larger format will affect frame rate)
- Hardware
- NVIDIA ORIN NX/AGX, Qualcomm 8550
- IR Cameras
- Boson+ with UVC
Boson SX8 with UVC
Neutrino LC with USB
Neutrino SX8 with USB
- Power
- <12 W (average)
<15 W (maximum)
- Video Output
- WebRTP, RTSP, and HDMI
Resources & Support
Assets below are shared by all models in this product family.
Prism A-ISR Datasheet
Integrating Teledyne FLIR OEM thermal camera modules is easier than ever with our library of how-to videos and more!
Know which camera specs match your application needs?
Filter and sort 110+ models in our camera selector.
Visit the Support Center for help with your Teledyne FLIR OEM products or submit a support ticket, by visiting our Technical Support Center.
Related Products
Prism AI
Object Detection and Tracking Software Libraries
- 40+ Detection Categories
- 4M Training Annotations
- Compatible with Qualcomm or NVIDIA
Prism Ground ISR
Ground Platform Object Detection and Tracking Software Stack
- Compatible with EO sensors, enabling multi sensor solutions
- Supports up to 15 object classes
- Accepts 14-bit IR data from Boson, Neutrino, and select third-party IR and EO sensors
Prism ISP
Advanced Computational Image Processing Software
- Super Resolution, MSX, VividIR, and more
- Supports LWIR, MWIR, and Visible
- Compatible with Qualcomm or NVIDIA
Prism SKR
Integrated Seeker and Autonomy Software
- Accepts Thermal and Visible Imagery
- WOSA-Compliant Interfaces
- Compatible with Qualcomm or NVIDIA