• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 770 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

IWR1642: Wildlife tracking

Paul Zwiers
Prodigy 20 points
Part Number: IWR1642
Other Parts Discussed in Thread: TIDEP-0100

Hello. I am a wildlife researcher and am interested in simultaneously following multiple individuals in the field. Traditional methods use RFID tags, but that is not an option as I require position information. It seems that the mmWave sensor might be able to help. The field is a complicated landscape (trees, branches, moving leaves and other animals), and individuals need to be identified as they move in and out of the sensor range (an individual that moves out of the sensor range must be identified as that same individual when it moves back into sensor range). Because of these issues/requirements, might there be a way to tag individuals that will either bounce back an identification number (similar to RFID tags), OR that will "respond" only to a given frequency (if the mmWave sensor is able to scan in multiple frequencies)?

  • 0
    TI__Guru 50240 points
    Hi Paul,

    Have you seen the People Counting demo? dev.ti.com/.../

    It does not track targets that leave a scene and then come back but this scene is actually very versatile and adjustable. I'd also recommend looking at the People Counting TI Design TIDEP-0100 www.ti.com/.../tidep-01000

    Cheers,
    Akash
  • 0 in reply to Akash Gondalia
    Prodigy 20 points
    I did, which is why I'm intrigued by the possibilities. The video though shows individuals in a very simplified environment. I am concerned that a large amount of non-target objects would be detected in a complicated environment (leaves and other moving things in addition to target individuals). It would be ideal to have something on the individual that would reflect a specific frequency, allowing me to tell the difference between target and non-target, and would identify the same individual as it moves in and out of range.
  • 0 in reply to Paul Zwiers
    TI__Expert 3610 points
    Hello Paul,
    Yes, the above scenario could be tackled with two step process.
    1. In the people counting and tracking application, the classification of objects is not implemented yet. With the classification implemented the problem of the moving things along with target individuals will be addressed. We will be able to separate out the target individuals or animals from the scene/background. This would also provide us with the location of these objects of interest.
    2. With RFID like device attached to the object of interest we may further identify them more precisely

    regards
    AK