How to mark specific points in a boundary?

I used this code to mark boundary of my image.Now I want to detect specific points such as most concave point in a specific region of the boundary and most convex point.I have given the image that i detect the boundary and i manually detect the points using blue arrows that i want to mark automatically.Can you please suggest a method for this?

I = imread('055.bmp');

 figure()

imshow(I);colormap gray

title('Original Image','fontsize',14);

img2 = imnoise(I,'salt & pepper',0.025);

I1 = img2;

for c = 1 : 3

    I1(:, :, c) = medfilt2(img2(:, :, c), [5, 5]);%add median filter

end

x =rgb2gray(I1);

I2 = x < 60;

 %I2 = im2bw(I1);

 figure()

imshow(I2);colormap gray

title('Median Filtererd','fontsize',14)

 dim = size(I2);

col = round(dim(2)/2)-90;

row = find(I2(:,col), 1 );

boundary = bwtraceboundary(I2,[row, col],'N');

%----Display original grayscale image and use coordinates returned by bwtraceboundary to plot the border on the image.

imshow(I);

hold on;

plot(boundary(:,2),boundary(:,1),'r','LineWidth',1);

BW_filled = imfill(I2,'holes');

boundaries = bwboundaries(BW_filled);

%------Plot the borders of the original grayscale image using the coordinates returned by bwboundaries-----------

for k=1:10

   b = boundaries{k};

   plot(b(:,2),b(:,1),'r','LineWidth',1);

end

Answer 1

[Image Analyst]

Try adapting this File Exchange code: https://www.mathworks.com/matlabcentral/fileexchange/32696-2d-line-curvature-and-normals

[Nayana R]

Thank you for your reply Image Analyst.But I don't know how to run a function in MatLab as I am new to MatLab.I tried it.I can't understand what are arguments they are asking as (Vertices,Lines). Can you please explain a way to run this function if you can?

In my previous code above edges traced are not saving in the image.How can I save image with edges and apply this function to my own image?I hope you can help me.Thank you.

Answer 2

[Jai Sri]

How to mark boundary of gray scale image in original color image?

[Image Analyst]

This does not look like an answer to Nayana's question. If you have a question, post a new question and post both your gray scale image and your original color image.