This script is a demo that shows you how to find objects of a certain color in an image. For this example I'm using the standard onion, peppers, or canoe image that ships with the Image Processing Toolbox. You can also select one of your own color images instead if you wish. Then I separate the image into its component red, green, and blue color bands. I compute thresholds, and ultimately find a mask for the red objects. Prompts will guide you through the demo step by step. I multiply this mask by the original image to show the image with only the red objects showing and everything else blacked out. This script can be adapted to find other colors (white, green, blue, yellow, etc.)
If you want to do any measurements of the red objects, I refer you to my blob measurement demo file at http://www.mathworks.com/matlabcentral/fileexchange/25157
Requires the Image Processing Toolbox. Tested with R2008b and R2010a.
Can I know what changes should I do in this code to extract dark colours(black shades)
Kaleb and others, I've uploaded a version that fixes the incompatibilities with the newer MATLAB releases.
I'm having the same issue as M Hq and am wondering if there was a solution? I've been looking but i cant see what index could exceed the dimensions or what matrices could be in question. Any help would be great.
It was helpful for me thank you
After the histogram part the code shows an error: Index exceeds matrix dimensions. No matter how big or small file size I choose this error exists. I wonder why?
Why does it show that you don't have IPT even though i have IPT. I've checked using ver too that i have IPT.
Anyways, after continuing it shows out of memory later. I've tried using matlabpool but it didn't have much effect.
Excellent demo .
tried using it on my images for my spectral analysis
Thanks for the detailed demo
Great for learning the thresholding and the masking of RGB and grayscale images.
Sir, in this demo, when i pick my own image some problem occurs.
1. When it computes histogram for blue, axes labels gets mirrored..!!! How to solve this.
(I am using Matlab R2012a)
In this, all three histograms are merged by line in a single subplot. Can we do it with bars also for the same means plotting all three histograms in same subplot with bars of individual color... how..??
Thank you for your help.
Thanks for a great demo! I used it with my own test image of some color bars to study the various methods you used. Thanks again.
Nice Work Sir!!
Can i know what changes should i do to extract only blue color?
excuse me sir, may i know how do you compute threshold?
Thank you Image Analyst for your help.
If anyone wants to ask me a question, click my name above, then click the "Contact Author" link.
i like your work
sir i want to detect two colors blue and red in the same image..i am not able two do that..please tell me how to do that..
Soy German Espinal, un estudiante de semestre 8 de la Universidad de Cundinamarca.
Actualmente estoy desarrollando un proyecto para la materia de Inteligencia Artificial y me urge un codigo Matlab para terminarlo.
Yo he encontrado muchos buenos videos y material sobre Matlab y creo que gracias a ellos he aprendido mucho.
Basicamente lo que necesito es:
Poner una tarjeta o un carton pintado de un color (puede ser primario o no) frente a la webcam de mi computadora
Capturar esta imagen si es necesario
Decir que color es ese : rojo, amarillo, verde, azul, blanco, negro, etc.
Parece simple verdad ? He buscado mucho, pero no se si debo capturar la imagen, hacerle algun tratamiento, trabajar en RGB o en que codigo de color, como capturar ese codigo y compararlo para poder dar el resultado.
Mil gracias !
Germán Espinal email@example.com
The Mathworks has a LAB color space classifier demo at
It's also worth looking at.
Fixed calls to stem() and surface() that were no longer compatible with release R2017b. However, it should work in all previous versions as well.
Now allows non-indexed monochrome images (simple gray level images with no stored colormap), and gracefully handles cases where there are no red blobs at all.
Expanded screenshot, and fixed small bug that happened when you were using your own image instead of demo images.
Added measurements of area, mean Red value, mean Green value, and mean Blue value for each red blob that was found.
Win cash prizes and have your live script featured on our websiteLearn more
Download apps, toolboxes, and other File Exchange content using Add-On Explorer in MATLAB.