I think that you can solve this problem by mathworks homepage.(www.mathworks.com)

On the homepage, we have a community for inquiries between users. 

using "find" command you can do it, let say for CbCr color space, you can filter a range of color like this:

[r,c,v] = find(Cb>=77 & Cb=133 & Cr

clear all

close all

clc

A=imread('Bag28.jpg');

[row,col,lay] = size(A);

A1=A(:,:,1);

A2=A(:,:,2);

A3=A(:,:,3);

Leave_redpixels = row*col-sum(A1(:)==255);

Leave_greenpixels = row*col-sum(A2(:)==255);

Leave_bluepixels = row*col-sum(A3(:)==255);

subplot(2,3,1),imshow(A1), title('Red Component');

subplot(2,3,4),imhist(A1);

xlabel('Greyscale Value')

ylabel('Number of Pixels')

title([num2str(Leave_redpixels),' Red Pixels'])

subplot(2,3,2),imshow(A2), title('Green Component');

subplot(2,3,5),imhist(A2);

xlabel('Greyscale Value')

ylabel('Number of Pixels')

title([num2str(Leave_greenpixels),' Green Pixels'])

subplot(2,3,3), imshow(A3), title('Blue Component');

subplot(2,3,6),imhist(A3);

xlabel('Greyscale Value')

ylabel('Number of Pixels')

title([num2str(Leave_bluepixels),' Blue Pixels'])

Thank you all, 

appreciate it.

I imagine that you have a range R_c = [min:max] where c is red, green and blue. Now for each image channel (R,G,B), you binarize the image such that pixels are set to one when they are within the range:

R_bin =  (R >= MinG) & (R = MinG) & (R = MinB) & (R