Sony Camera Parts count

  • Hello,

    I am dealing with 2D IRVISION to pick from multiple layer basket I have two Issues:

    1) with lower layer the camera vision will be not clear.

    2) this is my important point how can I detect how any parts exist by layer because now according to the current configuration the camera detect parts one by one, so do we have this option to let the camera scan all parts then start sending coordinate for parts to be picked one by one.



  • AD
  • Depth of field is just ability to focus and have a sharp image within a range. It is a physical setting on the camera lens and can't be changed dynamically, you set it once and never change it. It is controlled by the f stop or aperture size of the lens.

    Usually 8 works well for most applications. Going higher gives more depth but darkens the image.

  • You can focus on a mid layer, so the upper and lower layer aren't real sharp, but not as blurred as if the focus was on most upper layer.

    For the other thing: no idea.

    Edit: shouldn't be a problem, camera should make a picture and recognize next part at the same time when robot handles the picked part outside the pallet. No time lag.

  • How height are your layers, 1/4", 6", 2' ?

    How big is your FOV ? Can you see the whole layer, half, quarter ?

    I'm asking because:

    Depending on the height, I would have as many GPM as layer I have

    I'm assuming you can see the whole bottom layer. Can you see the whole top layer ?


  • Hello, now i am facing the same project as you.

    currently, i have 5 layers of workpieces that need to pick by the robot, every layer consists of 28 workpieces.

    the distance from the Camera lens to the bottom of the layer is around 1200mm in height.

    base on my experience, you can use this camera for your process, but there is something to concern about the below item.

    1. accuracy, higher the camera distance, it will decrease the capabilities of the camera regarding accuracy. If accuracy doesn't matter, please set your camera with the proper position, until all workpieces are detected and the focus also good. for the focus, you can set the bottom layer as the best focus, and let the rest follow the last focus setting. If you need to place your workpiece to the Jig directly with a precise pin and positioner, forget about this vision.

    2. programming, about this item, you will have easy teaching and programming using Depalletizing camera vision process. the height of the workpiece will automatically adjust by the different scales of the workpiece in each layer. but this setting only works 100% for flat workpieces in the flat base of each layer.

    3. calibration, please make sure to have a good result of the calibration. It is really impossible to get the best result only use the calibration grid. even if we use the 30mm calibration grid. i suggest using Robot generated grid calibration for this setting.

    4. lightning. please be attentive to the stable and good lightning position and intensity.

    - About how to find how many layers of the parts, you can use Depalletizing camera to detect the height of the workpiece and then change it by some logic.

    - For my case, I use the different GPM Locator program for every layer and then making a logic program.
    example :
    GPM Camera 1 for layer 1
    GPM Camera 2 for layer 2

    and then we will run the camera process, technically the GPM camera 2 will not detect layer number 1 and vice versa. It is happening because Layer number 1 and number 2 have different scales and different heights.

    Really appreciate every feedback, CMIIW.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account
Sign up for a new account in our community. It's easy!
Register a new account
Sign in
Already have an account? Sign in here.
Sign in Now