Ask Your Question

Madz's profile - activity

2018-05-30 16:26:53 -0500 received badge  Popular Question (source)
2016-08-02 11:18:26 -0500 received badge  Notable Question (source)
2014-12-02 10:18:17 -0500 received badge  Popular Question (source)
2013-04-15 02:09:49 -0500 asked a question Getting rid of tiled background

I'm using canny edges in a video stream, which captures objects on ground (the camera it self is moving). The problem I have is that the ground is tiled, therefore the tiled edges are captured as well. I did a Gaussian smoothing but it didn't remove the tiles.
Is there a background subtraction I can use to get rid of the tiles?


2013-04-06 05:11:57 -0500 asked a question Clear path detection using edge detection

I want to create an application where a user can detect an unobstructed path using an Android mobile phone. I used a Gaussian filter for smoothing and then used canny edge detection to detect edges.
I thought if I could find the concentration of edges in the video I can direct the user towards an area with a low concentration of edges, hence a clear path.
Is there a way to find the concentration of the edges? I'm new to image processing so any help would be appreciated.
Thank you

2013-04-06 05:07:12 -0500 received badge  Supporter (source)
2013-03-30 01:18:10 -0500 received badge  Editor (source)
2013-03-30 01:15:27 -0500 asked a question Missing opencv packages in OpenCV

I was going through some examples in I noticed that when I opened them in eclipse, some of the opencv packages were missing, such as;

  • com.opencv.jni
  • com.opencv.opengl

  • What is the reason for these packages to be missing in my openCV library? I googled for them and found them in this location.

    I use OpenCV Library -2.4.3. Thanks in advance.

    2013-03-28 08:57:20 -0500 commented question Android image processing and obstacle detection

    @jayrambhia Can you elaborate on how to achieve this with optical flow please?

    2013-03-27 01:17:48 -0500 asked a question Android image processing and obstacle detection

    Hi, I'm new to OpenCV and image processing. My objective is to detect obstacles via an android phone, the user carries the phone with him.
    I have used canny edge detection(which worked) and tried background subtraction(which didn't work) and am at a loss on what to do next. I need to detect an incoming object. What should I do next to detect an obstacle?
    Any help would be much appreciated.

    2013-03-10 07:23:44 -0500 received badge  Scholar (source)
    2013-03-05 01:34:01 -0500 received badge  Student (source)
    2013-01-28 09:45:55 -0500 asked a question NDK support for ARMv6


    I tried to run an openCV facial detection app on my Samsung s5360 which has a 2.3.6 android version. I keep getting INSTALL_FAILED_CPU_ABI_INCOMPATIBLE error message. After some digging up I came into the conclusion that the reason was the latest NDK revision 8d doesn't support ARMv6. Is there a suitable NDK version that would support my phone?

    Thanks in advance

    2013-01-27 23:25:33 -0500 asked a question Object tracking with android


    I want to create an android app where a live camera stream captures obstacles on the ground and issues warnings to the user. This is intended as an application for the visually impaired.

    I'm at a loss to understand what technologies are most suitable for this as I have no previous experience in image processing. I looked through some OpenCV examples, but they didn't provide me with much insight as to where to start the project from.

    What I need to know is;

    • Any tutorials that might help me to get started on the project (should I start with basic image processing?)
    • Should I learn C++, since I only know java and C#?
    • What technologies that could be used in the project. (Is it possible to achieve something like this through Augmented Reality etc..)
    • Any previous projects/ tutorials/ examples that tries to achieve android image processing

    If you could provide me with some answers to any of the above points I'd be much grateful.