Detect object from image based on object color by C#

To detect a object based on its color , there is an easy algorithm for that .u have to choose a filtering method . Steps normally are

  • Take the image
  • Apply ur filtering
  • Apply greyscalling
  • subtract background and get ur objects
  • find all the objects position
  • mark the objects

First u have to choose a filtering method , there are many filtering method provided for c#. mainly i prefer aforge filters , for this purpose they have few filter they are

  • ColorFiltering
  • ChannelFiltering
  • HSLFiltering
  • YCbCrFiltering
  • EuclideanColorFiltering

my favorite is EuclideanColorFiltering it is easy and simple. for other filtering u can know more about them here. U have to download Aforge dll for apply these in ur code .

So see EuclideanColorFiltering work first see

we are going to apply a color filter it is very simple code to use euclideanfiltering

// create filter
EuclideanColorFiltering filter = new EuclideanColorFiltering( );
// set center colol and radius
filter.CenterColor = Color.FromArgb( 215, 30, 30 );
filter.Radius = 100;
// apply the filter
filter.ApplyInPlace( image );

now see the effect

well to understand how it work look closely at the code

filter.CenterColor = Color.FromArgb( 215, 30, 30 );
filter.Radius = 100;

first line select the select color value. you all know color has a value 0 to 255.
by filter.CenterColor = Color.FromArgb( 215, 30, 30 ); i specified my center color will be a red effected color because here value of red is 215, green and blue is 30. and filter.Radius = 100 means that all color value near than 100 in my specified color.
now my filter filters pixels, which color is inside/outside of RGB sphere with specified center and radius – it keeps pixels with colors inside/outside of the specified sphere and fills the rest with specified color.

Now for detect objects i use bitmap data and use lockbits method to understand clearly this method see here . then we make it greyscale algorithom hten unlock it.

BitmapData objectsData = image.LockBits(new Rectangle(0, 0, image.Width, image.Height),ImageLockMode.ReadOnly, image.PixelFormat);
            // grayscaling
            UnmanagedImage grayImage = grayscaleFilter.Apply(new UnmanagedImage(objectsData));
            // unlock image
            image.UnlockBits(objectsData);

now the object we use blobcounter for that. it is a very strong class that aforge provided.

  blobCounter.MinWidth = 5;
            blobCounter.MinHeight = 5;
            blobCounter.FilterBlobs = true;
            blobCounter.ProcessImage(grayImage);
            Rectangle[] rects = blobCounter.GetObjectRectangles();
            foreach(Rectangle recs in rects)
            if (rects.Length > 0)
            {
                foreach (Rectangle objectRect in rects)
                {

                    Graphics g = Graphics.FromImage(image);

                    using (Pen pen = new Pen(Color.FromArgb(160, 255, 160), 5))
                    {
                        g.DrawRectangle(pen, objectRect);
                    }

                    g.Dispose();
                }

            }

blobCounter.MinWidth and blobCounter.MinHeight define the smallest size of the object in pixel.
and blobCounter.GetObjectRectangles() return all the objects rectangle position. and using graphics class i draw rectangle over the images.

now if u want to take only the biggest object there is a method for that

   blobCounter.MinWidth = 5;
            blobCounter.MinHeight = 5;
            blobCounter.FilterBlobs = true;
            blobCounter.ObjectsOrder = ObjectsOrder.Size;
 blobCounter.ProcessImage(grayImage);
            Rectangle[] rects = blobCounter.GetObjectRectangles();
            foreach(Rectangle recs in rects)
            if (rects.Length > 0)
            {
                   Rectangle objectRect = rects[0];
                    Graphics g = Graphics.FromImage(image);
                    using (Pen pen = new Pen(Color.FromArgb(160, 255, 160), 5))
                    {
                        g.DrawRectangle(pen, objectRect);
                    }
                    g.Dispose();
                           }

now image is like that

but if u want to extract u can use following code

Bitmap bmp = new Bitmap(rects[0].Width, rects[0].Height);
Graphics g = Graphics.FromImage(bmp);
g.DrawImage(c, 0, 0, rects[0], GraphicsUnit.Pixel);

so u will get ur object like that

now if u want to draw the rectangles in main image use that image reference in graphics. u will find u result. Hope this can be helpful

see update at this post

Advertisements

About kishor datta gupta

Graduate Research Assistant at University of Memphis Software Engineer at Silicon Orchard LTD. Former Research Assistant at Lamar University Former Software Engineer at Samsung R&D Institute Bangladesh Studies Ph.D. Computer Science at University of Memphis Studied Masters of Science in Computer Sciences at Lamar University Studied BSC in CSE at Khulna University of Engineering and Technology Studied HSC (completed) at Chittagang college 04-06 Studied High school at ST. Placid's High School'04 Studied Junior Secondary School at Saint Mary's School Lives in Memphis, Tennessee
This entry was posted in C#, Image Processing and tagged , , , , , , , . Bookmark the permalink.

16 Responses to Detect object from image based on object color by C#

  1. Sergio Gutiérrez says:

    Thanks man!!..your work is amazing!!

    This topic helped me a lot in my grade project!!…is a ANPR system…thank you very much!

    best regards.

  2. Anonymous says:

    hi,
    nice code examples.. currently,i’m doing my project on ear detection and i’ll try to adapt this code to my project later.. if possible,where can i find the full source code..?

  3. Anonymous says:

    thanks for your genius code; this code helps me a lot with my course project; but i’m doing right now my final project using c#: Quality Qualifier , this project uses image processing, how can we determine that the fruit mango be detected if it is a rejected, local, or maybe we can export this product, mango can be determine maybe by the color inorder this will be qualified for export, local, or rejected, …… thanks you very much sir;

    Christian Fletcher

  4. #trans thank you very much for sharing

  5. Thank you so much for posting this. It helps a lot in our projects…
    My only problem is how to extract the color of the object using RGB only. I only need the color of the object and extract it. Thanks again!

  6. jasmine says:

    this code helped me so much last year , why don’t you make another one based on features of the object not only the color ??

  7. Vinayak says:

    hello Kishor D Gupta
    i want to know is der anyway to detect two colors in d same way?
    i tried using two blob counters but its giving an unmanaged code error.
    can u please help! 🙂

  8. gaurav says:

    thanks

  9. Ibrahim says:

    Hi, I am Ibrahim, a masters student of university of east London. I am currently writing my project on object detection from a 3D image. can you be of assistance to me?

  10. lenny lemor says:

    Hi, great code,
    how resolve the error at this line: filter.CenterColor = Color.FromArgb(0, 0, 0); (line 156 about)
    ‘implicit conversion System,Drawing.Color in Aforge.Imaging.RGB’
    I use the last release of Aforge lib.
    Thank in advance.

    • Anonymous says:

      Hi,
      u need to do it this way :
      Color pixel = your_bitmap.get_pixel(x,y);
      Aforge.imaging.RGB your_color = new RGB(pixel.R,pixel.G,pixel.B,pixel.A);
      filter.centercolor = your_color

      and it works 🙂

  11. ilyes says:

    code that doesn’t compile !

  12. Richie says:

    thanks for the help i am currently having a thesis on meat classificatiob

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s