Machine Vision

cancel
Showing results for 
Search instead for 
Did you mean: 

How to get state information from cameras

Hello, I am very very new to vision based estimation (actually vision based anything), but I want to be able to gather trajectory state information of an object from two cameras. One camera to get the XY coordinate (placed overhead) and another for XZ coordinate. I want to know the hardware and software I will need to accomplish this. I would like to gather the telemetry in realtime for state estimation processing. The state estimation part understand but the data acquisition is new to me. My background is in theory and simulation and I no experience in real hardware and software. I am assuming I will need a NI Vision Acquisition Software and two cameras.

Any advise is appreciated. I just don't know where to start.
0 Kudos
Message 1 of 2
(4,002 Views)

Hi there,

 

You can track objects with just one camera, two cameras are used for analyzing the depth of an object (unless necessary for your application, in which case I would recommend checking this webcast: 3D Vision and Kinect).

 

For telemetry, here is another resource that can help: Making Measurements on the Image. Since you are getting started check the “Table of Contents” link for more information and key concepts.

 

Also, there are a lot of examples out there that can get you started, for example:

 

- Community: Googly Eyes (Powered by LabVIEW)
- object tracking using labview using camera

 

Last but not least, the software components that you need from NI are: LabVIEW (development environment), Vision Development Module (add-on for vision functions) and Vision Acquisition Software (driver that talks to cameras). You can download this software from the NI website and use it in evaluation mode. If you are interested in the licensing, here is more information: Licensing National Instruments Vision Software

 

I hope this helps to get you started

Alejandro C. | National Instruments
0 Kudos
Message 2 of 2
(3,962 Views)