Pensar Camera + SDK

  • $16,402.00
Aerialtronics PENSAR CAMERA & SDK
Develop and run your AI applications
Need to develop AI application and process real-time data? Create your own AI apps from concept to deployment within a few days with the AI-powered Pensar camera and SDK. Pensar is the fastest, most power-efficient embedded AI computing device: NVIDIA Jetson TX2. Pensar is an AI-powered camera dedicated for professionals and integrates 
a Sony 30x zoom HD camera, an IR Flir Boson and an Nvidia Jetson TX2 GPU
WHAT IS PENSAR SDK? Pensar SDK is an end-to-end solution for AI application development which includes a Pensar Camera, an API, a dashboard and an SDK. Develop on PC and deploy on Pensar camera
  • Application and GUI development based on Python on top of C++ hardware-accelerated libraries.
  • Use the available pre-trained neural networks or import neural network in most common frameworks caffe, darknet and soon tensorrt)
Requirements for developers ‣ Programming languages: Python 3, C++
‣ OS: Linux based operating system (Ubuntu, Mint…)


  • Edge computing (no cloud dependency)
  • Real-time processing: no cloud latency
  • Reduce development costs
  • Simpler and faster developing time
  • Versatility – Cross-sector applications
  • Modular SDK
  • Low power consumption


Facial recognition

  • Multiple faces detection and recognition
  • Multiple faces annotations
  • Detected faces can be saved to a permanent database
  • An external faces database can be loaded for recognition
  • Faces to be detected can be sent to Pensar from an external device (i.e. mobile phone)

Object detection

  • 80+ objects supported (person, animals, means of transportations, objects of common use, food, plants, etc.)
  • Several kinds of Neural Networks available, with different level of performances
  • Specific actions can be programmed when selected objects are detected (i.e. take a geotagged picture, etc.)

Car license plate recognition

  • Uses a combination of two neural networks, the first one to detect cars (other vehicles can be added in future) and the second one to detect and recognize car registration plates
  • Optical character recognition is used to extract license number plate. Performances of recognition depend on optimizations specific for each country

Optical Tracking

  • Any image area can be chosen and tracked on-demand by the user
  • Different tracking algorithms can be used in order to get best results on given scenario
  • Gimbal (if present) is controlled by Pensar to perform physical tracking
  • Supports both optical and thermal camera

AI Powered Object Tracking

  • AI is used to automatically detect objects to track (person, car, train, …)
  • User can choose which object to track amongst detected ones
  • Gimbal (if present) is controlled by Pensar to perform physical tracking
  • 30x optical zoom is used to track objects on z axis. Object size is automatically kept constant on screen. Maximum tracking distance up to 200 meters (it depends on object type and size)
  • Limited support for thermal camera but specific neural network can be trained and added in the future 

Thermal detection & Tracking

  • Object detection based on thermal imaging (see “Optical detection” above)


 Video output digital HDMI
Data Storage 64GB Micro SD card *16GB eMMC internal storage
Connectivity  WiFi/Bluetooth/Micro USB
Voltage  24V



Resolution  1920x1080/30fps
Optical zoom 30x
Image device  1/2.8-type CMOS
Horizontal viewing angle 63.7°(wide) to 2.3°(tele)
Focus system  Auto/Manual
Min. object distance** 

10 mm to 1200 mm
**Default: 300 mm

Electronic shutter  1/1s to 1/10,000s
Focal length
(35mm converted)
33.0 mm to 659 mm



Resolution 320x256
Pixel pitch 12 µm
Sensor technology  Uncooled VOx
Thermal sensitivity < 40 mK to < 60 mK
Spectral range 7.5 - 13.5 µm
Digital zoom  Yes
Slow frame rates  < 9 Hz available
Focal length  13.8 mm



Outer dimensions  112 x 98 x 67 mm
Gimbal AV130 & AV200
Mechanical interface 1/4-20 screw thread
Weight 672 grams



GPU:   NVIDIA Maxwell™, 256 CUDA cores
CPU:  Quad ARM® A57/2 MB L2

4GB 64 bit LPDDR4 25.6 GB/s


*Using the “Linux4Tegra” (L4T) tools, it is possible to back-up and restore Pensar images.


Pensar software stack

Pensar is provided with: standard Linux OS, middleware of hardware optimized libraries, proprietary AI SDK (Plugin Manager) and several ready-to-use AI applications. In general, the software stack provided with Pensar is:

Linux OS and Middleware

Operative system currently used is Ubuntu 16.04, it includes CUDA 9.0 accelerated drivers for NVIDIA GPU. Middleware is constituted by a large set of libraries, all available as open source, a big effort has been spent to integrate them on Pensar taking maximum advantage from hardware acceleration.

AI hardware accelerated libraries provided with Pensar middleware include:

● NVIDIA cuDnn 7

● NVIDIA TensorRT 4

● NVIDIA Caffe

● OpenCV 3.4.5 (CUDA and ARM Neon acceleration supported)

● DLIB (CUDA and ARM Neon acceleration supported)

● GStreamer with NVIDIA proprietary CUDA accelerated plugins

● Darknet (CUDA accelerated)

TensorRT, Caffè, OpenCV, DLIB and Darknet make possible to load and run the most common AI neural network model formats that include:

● TensorFlow

● Caffe


● Darknet networks (like for instance YOLO neural networks)

Other libraries provided with Pensar middleware include:

● Keras

● Theano

● TensorFlow

● SciKit

● SciPy

● Numpy

● Matplotlib


● Mavlink

● Kivy (used for Graphical User Interfaces)

● Docker-ce

● Visual Studio Code

Python bindings have been included for all C++/C libraries (if exist), so in general the middleware make possible to write hardware accelerated applications in C++ and Python.

App Manager

On top of that there is App†Manager. App Manager is a multiplatform software entirely designed in Aerialtronics, written in C++ and Python, it is not an application but an AI and Computer Vision environment to build and run generic applications.

App Manager includes the following functionalities:

● basic Pensar control and Graphical User Interface to handle sensors (visible camera, thermal camera, IMU etc.) and other Pensar hardware

● basic functionalities (take snapshots, record video, etc.)

● capability to communicate and control all other hardware connected to Pensar (operate Gimbal to track targets, talk to fly computer, etc.) making use of Mavlink, IP and USB connections

● capability to run generic applications that have their own Graphical User Interface, fully integrated into Plugin Manager GUI (Kivy is used to get OpenGL accelerated GUIs)

Applications can be written in Python, are made of a sequence of plugins (that we call plugin pipeline) and, thanks to Plugin Manager, can easily access to AI neural networks running in C++, making use of hardware accelerated middleware libraries. Some neural network models are provided with Plugin Manager, others can be loaded since Plugin Manager supports most common neural network standards.

    If you are interested in Renting this product checkout our Rental website at

    We work with 2 different companies for financing.  Each Organization has their own criteria for purchases.

     BNC Financial
    Specializes in equipment funding from $500 and up. Follow this link to to the the financing application.

     First Pacific Funding
    Financing options from $5000 and up. Follow this link to go to the financing application.

    Need Support?

    Visit Here