Build a Parking Space Recognition system with Computer Vision

Поделиться
HTML-код
  • Опубликовано: 10 дек 2024
  • Build your own AI vision solutions: pysource.com/c...
    ➤ PROJECT SOURCE CODE:
    The source code for this project and a full tutorial where I show step by step how to build this project from scratch is available to download inside the course Object Detection (with Opencv and Deep Learning) here
    pysource.com/o...
    In today's video we're going to talk about how to build a Parking space Recognition system that works in a real world scenario, what technology to use and we'll focus on the main phases of this project.
    ➤ Courses:
    Object Detection (with Opencv and Deep Learning): pysource.com/o....
    Training Mask R-CNN PRO (Notebook + Mini-Course): pysource.com/m....
    Blueprint: pysource.com/b...
    ➤ Follow me on:
    LinkedIn: / pyso. .
    ➤ For business inquiries:
    pysource.com/c...

Комментарии • 19

  • @pysource-com
    @pysource-com  2 года назад +1

    🔥Learn how to build your own AI vision solutions: pysource.com/community

  • @ecsyntric
    @ecsyntric Год назад

    5:06 good point. many newbies claim they can detect parking spots from security cameras. these cams usually are at low angle and impossible to distinctly see each spot.

  • @ronakttawde
    @ronakttawde 2 года назад +1

    Awsm video expalination.
    One Suggestion: You can ADD on to this project stating that how many parking lots are "OCCUPIED" and "VACANT" by adding a counter by color coding to parking space in real time. further more we can allot the specific parking space to achieve FIFO

  • @bolzanoitaly8360
    @bolzanoitaly8360 2 года назад +1

    Wonderful idea, I really appreciate your open mind research and teaching to all of us.
    We appreciate you and your wonderful videos.
    thanks, a lot.
    you are the master in Computer Vision.
    best regards
    Gul Rukh

  • @MrFudgeYT
    @MrFudgeYT 2 года назад +2

    I really love the way you explain every step. Thank you so much

  • @obi666
    @obi666 2 года назад +2

    Hello, is it possible to share video from 10:00?

  • @coccosapiens
    @coccosapiens Год назад +1

    Hi Sergio! Thanks as always for your great work... My question is: which kind of cameras do you recommend to do this project? Since you said in the beginning that usually the quality of common CCTV cameras is not good. Is there any suitable CCTV camera for computer vision purpose? thanks

  • @jaderrosa2
    @jaderrosa2 2 года назад +1

    Really good video Sergio. As a suggestion I really like to see some video / material about parts measurement with high precision and how to define what will be accurate in millimeters according to the quality of the image acquired

    • @pysource-com
      @pysource-com  2 года назад

      Hi Jader, could you please elaborate more this specific request, giving me some practical examples on where you would use this type of software?

    • @jaderrosa2
      @jaderrosa2 2 года назад

      @@pysource-com imagine a rectangle of 2 meters long by 10 centimeters wide passing on a conveyor belt always in the same position. In this case the system needs to make sure that the part is 2 meters long by 10 centimeters wide with a tolerance of 0.5 millimeters. If out of tolerance, take some action

    • @pysource-com
      @pysource-com  2 года назад

      @@jaderrosa2 I don't know about this particular specific project as it would be necessary to make a research and test the accuracy with industrial cameras. Especially I'm not sure about whether it's possible to achieve a 0.5mm precision (unless the camera is Just a few centimeters away from the target.

    • @antonioconsiglio95
      @antonioconsiglio95 2 года назад

      @@jaderrosa2 it's possibile with the correct HD. You have to ensure that 0,5/(resolution mm/pixel) should be at least equals to 10. And morover avoid blurred edgees

    • @leo1722467
      @leo1722467 2 года назад

      @@jaderrosa2 maybe if you put your camera a little higher than the belt and define an área in Px a little bigger than the rectangle, then when the rectangle achieve the right spot on bottom you take a crop of the defined area that the rectangle should be and divide the mm/Px to know how much Px means 1mm for example, then you check how much of area inside your crop is filled with your rectangle.
      Hope you understand what I tried to transmit.

  • @mk-xx1vj
    @mk-xx1vj 2 года назад

    this is really nice. thank you :)) Can you show us how to measure the width, height and depth of the shipping box with any stereo camera?

    • @pysource-com
      @pysource-com  2 года назад +2

      thanks for the suggestion. I'll take note for this project for future videos, as measurement through computer vision is quite requested

  • @shyampatel7784
    @shyampatel7784 Год назад

    Hello sir, i am a college student and i am planning to build final year project, i have planned to build parking management system with booking and navigation in laravel and i want to integrate this open cv model, can i integrate this in laravel