Figuring out which oblejct is placed on pressure sensitive mat

Hi! I'm currently doing a project using a DIY pressure sensitive mat which is the same as the one in the followong link: http://www.instructables.com/id/O-mat/ I have just copied all the code from this tutorial and it is working as in this video:

I want to use this mat to give me a live update of how many products that are placed on the mat and also determine if its tin cans or lunch boxes. I'm thinking that this might be possible to do with some kind of image recognition but since I am a total beginner in both C++ and Java (I know Matlab and a little bit of Phyton but i guess that cant help me here) I dont really know where to start and what I need to do.

I will be able work on this project for about 20-40 hours more so anything from working code to ideas or a good read will be very appreciated.

Where should i start?

Answers

  • Thanks! I have been doing some research so now i wonder if im on the right track if i would base my program on this code: https://docs.opencv.org/2.4/doc/tutorials/objdetect/cascade_classifier/cascade_classifier.html#cascade-classifier

    and create my own xml-file using this method:

    Do you think i will be able to do all of this in Processing or will there be a problem?

  • I can't play the last video but it is probly bc of my current network. Nevertheless, the code above is in C++. You should look into a java version first. Maybe the library accepts an xml format directly or it will have its own input requirements. Notice Processing has functions that can handle xml data.

    Also explore the OpenCV library that you can install directly from the PDE. When you install the library, you can explore the examples that come with it via Files>>Examples>>Contributed libraries.

    Kf

  • Alright, thanks!

    I found this and think that this might work as a foundation for my project https://docs.opencv.org/2.4/doc/tutorials/objdetect/cascade_classifier/cascade_classifier.html#cascade-classifier

    This code uses face recognition on the video from a web cam. I guess the changes I need to do is create new .xml files and change the input to screen caputure instead of web cam video.

    The problem is that i dont know how to do this. Is it possible to create a second "window" that "looks at a window that I have already created using Processing?

    Right now it looks like this (from the link above):

    //-- 2. Read the video stream capture = cvCaptureFromCAM( -1 ); if( capture ) { while( true ) { frame = cvQueryFrame( capture );

    So i guess what I have to change is "cvCaptureFromCAM( -1 )" to something else?

Sign In or Register to comment.