Aerial Object Tracking
C++ and OpenCV prototypes exploring the detection and tracking of selected objects in aerial imagery.
Computer-vision prototypes and concept documentation
My contribution
Computer-vision prototypes and concept documentation
Team & credits
- Hossein Niromandi — Co-inventor
Delivered scope
Computer-vision prototypes and concept documentation forming a foundation for further embedded-vision research.
What stands out.
- Moving & selected-object prototypes
- C++ vision implementation
- Collaborative invention documentation
From the problem to the implementation.
The problem
Aerial imagery can support more autonomous sensing when relevant objects can be identified and tracked over time.
The approach
- Explored detection and tracking of moving and selected stationary objects.
- Used C++ and OpenCV to process aerial-video imagery.
- Contributed to invention documentation connecting visual information with a UAV-control concept.
Result & scope
Computer-vision prototypes and concept documentation forming a foundation for further embedded-vision research.
Implementation choices and demonstrated scope.
Engineering decisions
The C++/OpenCV desktop prototypes process recorded aerial and road footage, with operator selection, region detection and adjustable area, threshold and morphology controls. The separate collaborative concept connects extracted visual information with possible onboard UAV decisions.
Validation & evidence
Six archival desktop recordings show vehicle-region detection, selected-target tracking and adjustable image-processing controls over recorded footage. The cover is a labelled AI concept illustration. These artifacts demonstrate desktop prototypes; they do not establish an onboard UAV implementation, flight validation or measured tracking accuracy.
Explore 1 project images and 6 archival recordings, with captions identifying the hardware, interface or prototype shown.
Inside the project
1 image
Enlarge imageSee it in action
Vehicle tracking over desert terrain
A desktop OpenCV prototype follows a vehicle in desert footage, with a tracking box and adjustable area, threshold and morphology controls. Archival desktop recording; silent playback.
Open videoHighway detection and selected-target tracking
The desktop prototype switches between vehicle-region detection and following a selected vehicle in overhead highway footage. Archival desktop recording; silent playback.
Open videoSelecting and following a single vehicle
A target-selection demonstration follows one car along a road, with a green tracking rectangle and adjustable image-processing controls. Archival desktop recording; silent playback.
Open videoVehicle-region detection in traffic footage
Area annotations mark moving vehicle regions in a road-video demonstration, with area and threshold settings shown alongside the C++/OpenCV prototype. Archival desktop recording; silent playback.
Open videoHighway detection and region controls
Desktop vehicle-region detection over recorded highway footage, with red area annotations and adjustable processing controls. Archival desktop recording; silent playback.
Open videoSelecting vehicles in highway footage
Operator selection and green-box tracking of vehicles in recorded highway footage. Archival desktop prototype; silent playback.
Open videoHow the system works
Select a block to explore its role.
A guided illustration of the system components.
Aerial imagery
Video frames provide the input to the prototype.
Object selection/detection
C++/OpenCV work identifies moving or selected stationary objects.
Tracking
Visual information is followed over time.
UAV-control concept
A conceptual link from extracted visual information to onboard UAV decisions.
The detail behind the build
Detection and tracking in aerial imagery
C++/OpenCV prototypes explore moving objects and selected stationary objects in aerial video, connecting computer vision with autonomous sensing.
Visual information for more autonomous decisions
The project concept considers how extracted visual information could support onboard UAV decisions and reduce dependence on a human operator and continuous ground communication.
A collaborative concept
The invention documentation names Hossein Niromandi and Sina Shiri. The work combines image processing, electronics and machine vision in an aerial-sensing context.
Software, hardware & tools.
Software
- C++ computer-vision prototypes
Technology stack
- Computer vision
- Object tracking
- C++
Project contributions.
- Hossein Niromandi
- Co-inventor
Working on a related challenge?
Get in touch about Python, embedded systems or a research opportunity.
Industrial DataGraph
Embedded systems