• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 708 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

IWR1642BOOST: IWR1642BOOST Matlab processing People Counting Industrial Toolbox 2.5.2

Vyacheslav Kobzev
Prodigy 65 points
Part Number: IWR1642BOOST
Other Parts Discussed in Thread: IWR1642

Hello! I am developing an application that would allow working with a IWR without a Matlab graphical application. After I developed a parser for processing data from UART based on the well-known protocol for People Counting, I visually compared the work of my application and the Matlab-based GUI program, after which I got the impression that the output of UART was not enough to identify people in room. My program does not provide any processing of the output data from the device, and I think that Matlab processes the data separately to increase accuracy. At the same time I met in the sources of TI that all data processing is performed on the IWR device. I am not an expert Matlab and it is difficult for me to analyze the source code of People Counting in order to determine which individual software modules I should implement in order to increase the accuracy of the definition of people.

  • 0
    TI__Mastermind 28055 points

    Hi Vyacheslav,

    All of the People Counting processing is done one the device. The IWR1642 is outputting the location of each person - the Matlab GUI is just displaying these locations.  It is possible that you will need to tune the tracker to improve performance. Can you describe the inaccuracies that you are experiencing?

    Regards,

    Justin

  • 0 in reply to Justin Curewitz
    Prodigy 65 points

    Hello, Justin! I apologize for the long answer - the fact is that I just figured out my problem and as it turned out it was in my code, now everything works quite well, we are doing research. However, Justin, tell me, do you know anything about the effect of vibration of the IWR1642 on the occurrence of ghostly targets? We checked it and with fluctuations similar to those of the motor vehicle we had false alarms. Can you advise anything, how can you get rid of it? The fact is that we want to place the IWR1642 on an industrial combine for emergency shutdown in case of a person being found nearby during operation.

  • 0 in reply to Vyacheslav Kobzev
    TI__Mastermind 28055 points

    Hi Vyacheslav,

    I have not considered this use case before and will have to look into the issue of vibrations on the EVM causing false detection. 

    Please open a new thread as this is a different topic then the original thread. By having one topic per thread, it is easier to find answers.

    Regards,

    Justin