MATLAB: Projecting vision application for industrial robot

Question

Hello. I have task for project. I need create application in Matlab into a robot that will perform pick-and-place tasks. Can someone please advise me whether I think in the right direction. Preliminary scheme looks like this:

1. 1. Taking a photo
2. 2. Counting how many objects and where they are located
3. 3. I recognize the object and classifies it as such. for example Triangle or diamond.
4. 4. Downloading its orientation in (x, y) and send to the robot.
5. 5. The robot will have a separate program (pick and place) running in a loop.

For now, what i have in my code.

1. 1. Taking a picture.
2. 2. Converting to binary
3. 3. Filling gaps (holes).
4. 4. Labelling.
5. 5. Using region props witch can determine their orientation, middle.
6. 6. Now I want that every object is rotated to angle 0 '. (the same as uploaded template).
7. 7. Using the template matching check whether the object agrees with the template.
8. 8. When fits send information to the robot.

I'm not good with programming. I hope that someone will give me some tips. Please help.

My code here:

read the second image

  img2 = imread('https://s14.postimg.org/y3pheo535/rozsypanka.jpg');
  img2 = im2double(rgb2gray(img2));
bw = im2bw(img2); % Otsu's thresholding
bw = imfill(~bw, 'holes');                              % fill holes
figure(9),imshow(img2)
% show centers and bounding boxes of each connected component
L = bwlabel(bw);
figure(1),imshow(L == 7)
title('Object 7')
figure(9),imshow(L)
% Get the extrema points for each labeled object.
s = regionprops(L, {'Extrema' 'Centroid', 'BoundingBox', 'Orientation'});
imageHandle = imshow(L, 'InitialMagnification', 'fit');
axesHandle = ancestor(imageHandle, 'axes');
figure(2),imshow(L)
hold on
for k = 1:numel(s)
    c = s(k).Centroid;
    text(c(1), c(2), sprintf('%d', k), ...
        'HorizontalAlignment', 'center', ...
        'VerticalAlignment', 'middle');
end
extremas = cat(1, s.Extrema);
%hold on
plot(extremas(:,1),extremas(:,2), 'b*')
figure(3)
hold off
%Makes bounding boxes and X in centers of images
centers = vertcat(s.Centroid);
imshow(bw), hold on
plot(centers(:,1), centers(:,2), 'LineStyle','none', ...
    'Marker','x', 'MarkerSize',20, 'Color','r', 'LineWidth',3)
for k=1:numel(s)
    rectangle('Position',s(k).BoundingBox, ...
        'EdgeColor','g', 'LineWidth',3)
end
hold off
%Rotations
%c = s(r).Centroid;
%It can be done it loop:
%for (L == 1):( L== 7)
Object1=(L == 1);
Object2=(L == 2);
Object3=(L == 3);
Object4=(L == 4);
Object5=(L == 5);
Object6=(L == 6);
Object7=(L == 7);
st = regionprops( Object7, 'Orientation' )
rbw = imrotate(Object7, st.Orientation );
pc = any( rbw, 2 ); %// project all rows into a single column 
pr = any( rbw, 1 ); %// project all columns into a single row
fx = find( pr, 1, 'first');  %// first x coordinate
tx = find( pr, 1, 'last');   %// last x coordinat
fy = find( pc, 1, 'first');  %// first y coordinate
ty = find( pc, 1, 'last');   %// last y coordinate
imshow(rbw,[],'border','tight');
hold on; 
plot( [fx tx tx fx fx], [fy fy ty ty fy], ':r', 'LineWidth',3);
[fx fy tx ty]
st1 = regionprops( rbw, 'Orientation' );

Answer 1

See my shape detection demo, attached.

You might also want to look at Hu's moments for scale invariant, rotation invariant recognition: http://www.youtube.com/watch?v=Nc06tlZAv_Q