+ Post New Thread
Results 1 to 3 of 3
  1. #1
    Newbie level 1
    Points: 17, Level: 1

    Join Date
    Aug 2018
    Posts
    1
    Helped
    0 / 0
    Points
    17
    Level
    1

    Image processing for self drive car

    Hi experts!
    this is my first post in this forum. thanks edaboard

    I am final year engineering student. My final year project is Self drive car in which i will use Raspberry pi development board.
    My flow char for this project is:

    capture image from front camera ==> lane detection using python==> steer the car in between the lines

    I think i will successfully perform first two steps. but i am really confused in third step.
    How i will transform image into a signal which steers the car motors to remain in center of the road lanes either the road is curved or straight.
    Please help me to find the solution. Thank you so much

    •   AltAdvertisement

        
       

  2. #2
    Super Moderator
    Points: 77,851, Level: 68
    Achievements:
    7 years registered
    Awards:
    Most Frequent Poster 3rd Helpful Member

    Join Date
    Apr 2014
    Posts
    15,774
    Helped
    3592 / 3592
    Points
    77,851
    Level
    68

    Re: Image processing for self drive car

    Hi,

    Do a test with two humans on a closed road:
    * one person simulates the camera and the lane detection
    * the other person is steering the car

    Blindfold the person who is steering the car.
    He must just perform your commands.
    "Left", "right" ....


    ****
    The same needs to happen with your application:
    The python software needs to output similar commands. "Left", "right" ....

    And a motor(controller) needs to receive the commands and control the steering wheel.

    Either you can use the built in motor for servo controlled steering or you must install an additional motor.


    Klaus
    Please don´t contact me via PM, because there is no time to respond to them. No friend requests. Thank you.

    •   AltAdvertisement

        
       

  3. #3
    Super Moderator
    Points: 81,434, Level: 69
    Achievements:
    7 years registered
    Awards:
    2nd Helpful Member
    betwixt's Avatar
    Join Date
    Jul 2009
    Location
    Aberdyfi, West Wales, UK
    Posts
    13,349
    Helped
    4456 / 4456
    Points
    81,434
    Level
    69

    Re: Image processing for self drive car

    On a technical level, you need to find where the lines are in the image and then find the center point between them. If the center of the image is to the left of the center of the lines, steer right, if it's to the right, steer left. The critical points are the center of the camera image (assuming it looks along the vehicle axis) and the center point of the two lines.

    You need to do some image analysis to find the center of the lines, the image center is just half the horizontal camera resolution. As the image is mapped to video memory space, it may be possible to scan addresses corresponding to a horizontal line in the picture to find the two lighter points or you could do it by sampling brightness at a set of points within the image to find them. It would be wise to scan several horizontal lines, the danger of scanning a single one is it can only find gradual bends in the path, if for example there was an abrupt turn, one of the lines could fall outside the image area and the center calculation would fail.

    Brian.
    PLEASE - no friends requests or private emails, I simply don't have time to reply to them all.
    It's better to share your questions and answers on Edaboard so we can all benefit from each others experiences.

--[[ ]]--