I'm just beginning with Image analysis in MATLAB.
My goal is to do an automated image segmention on images of plant leaves.
I have had reasonable success here thanks to multiple online resources.
The current objective, the reason why I'm placing this question here, is to able to place 25 equidistant points along each half of the margin/outline of leaf, like described in following image:
For the script to be able to recognize each half of the leaf, user can put two points within the GUI. One of these user-defined points will be on the base of leaf and the other on tip of leaf. It would be even better if a script would be able to automatically recognize these two features of the leaf.
For the output, I would like a plain text format file containing image coordinate of each point.
I'm not asking for a ready made script here, but looking for a starting point.
One way I think this can be done is by linearizing/open up the outline in such a way that it becomes a straight line. This can be done by treating any of user placed point/landmark as breakpoint. Once a linear outline is obtained it can again be broken into two halves at other user defined point and now points can be placed. One point to bear in mind here is that the placement of points for each half should start from the end that corresponds to the same breakpoint/user-defined point in each half. Now these straight lines can be superimposed on original image for reconstruction.
Thank you very much.
Parashar
Related
I am currently in High School, and I am in an APCSP (AP Computer Science Principles) class, which in my case is learning in Scratch programming. I am confused and have practically no idea what I'm doing. Scratch is very confusing and I feel like it's pointless to learn.
My question is this: Can anyone help me on how to make a Maze Generator on Scratch, as this is my project and it's giving me struggles.
Thank you.
It's actually possible to build with scratch but depends on what you are looking for. I assume you want to generate a simple maze like in old fashioned 8-bit games like boulder dash.
First decide on the size of your maze: for example 5 x 5 blocks.
If you want to create a maze, imagine drawing it on a grid on paper. Blocks are either "empty" or filled in. Our maze can be represented by numbers. The empty blocks are represented by a 0 and the filled blocks with a 1.
You could visualize that matrix like this if all blocks are empty:
0,0,0,0,0,
0,0,0,0,0,
0,0,0,0,0,
0,0,0,0,0,
0,0,0,0,0
Adding a border wall while keeping the inside empty would look like:
1,1,1,1,1,
1,0,0,0,1,
1,0,0,0,1,
1,0,0,0,1,
1,1,1,1,1
Using a "list" variable to store this information would fit best within the possibilities of MIT Scratch.
In this case, you need to understand that each block in our maze is represented by a position in above matrix. You could draw numbers on a piece of paper in the shape and size of your grid / matrix as a reference to remember the position of each block if that makes it easier.
We also need to look at how our maze will relate to the Stage size. The width and height in pixels of a default scratch project is 480x360.
A 5 x 5 maze is divided in blocks of 480 / 5 = 96 width and 360 / 5 = 72 height. In other words, a block needs to be 96x72 pixels, based on a full screen maze.
Next step, is creating a sprite representing the visualization of the blocks of the maze. I would keep the first "costume" of our block sprite empty, and create a fully filled block to represent the walls of the maze.
After that, we need to programmatically create our maze. I made an example you can explore of random drawing of the blocks of a maze:
https://scratch.mit.edu/projects/278731659/
(You can change the rows & columns value to see it scale up, but remember the limit to the amount of clones the block sprite can have is 300)
This is just to get you started and by no means a complete solution. I just hope this helps you think in the right direction.
You can make this more advanced, by adding a function to explore and correct our randomly drawn grid to generate a walkable path from position x to position y. A rule you can program is for example: Every empty position in the grid should have at least two other empty positions in the spaces above, below, left and right of it.
There are many different ways to do this; whether this is with sprites and stamp or 2D lists and pen. Either way, the main component is the algorithm. This wikipedia page gives details on how maze generation works and few different algorithms. There is also a video series by The Coding Train here where he creates a maze generator with the 2D list method from above (this method is a bit harder on scratch, however). Either way, the best thing to do is to look at examples others have made, figure out how they work, and try to recreate them or make them better. Here's a good place to get started.
Scratch IS truly pointless! A simple maze generator would have you use the pen to draw predefined shapes (Such as a long hallway or intersection). You should also make (invisible) squares to separate everything and have the program draw in the squares.
I will put a link later that leads to a sample project that has the code.
Check out this video by griffpatch
https://www.youtube.com/watch?v=22Dpi5e9uz8
This was one of my projects, and the instructor provided this video for everyone to follow and expand from.
I'm working on MATLAB on some regions inside an image. I'm at a point in which I would like to be able to separate regions which exhibit some kind of regularity (e.g., being circle-ish or square-ish) from regions which does not resemble any known figure and which for my application are mere noise. I'll illustrate this using a descriptive MS Paint image:
Is there any tool that, most of the times (or even less, I know this can't be 100/100) will recognize the red thing as being different?
I'll deal with many shapes in a single image, so I don't mind if I carry on some red monsters along the way, as long as the majority of them is kicked out. Of course I know the indices of these regions, so I can manipulate them in MATLAB.
Many algorithms come to mind, e.g., getting the boundary and checking for its regularity/the number of times it changes curvature/..., checking for variations in vertical length through different columns (nearly 0 for the linear feature, really high for the red stuff), ...
However I was hoping in some help from a tool out there. It doesn't matter if this tool won't cover all cases (for example, will kick out circles), I've been very broad to get the maximum number of inputs from you guys - any tool will be inspiring and helpful (and, however, we can't expect a perfect answer for the deeper question - recognizing regular shapes - which seems more like a AI field of research). I also think that, while being broad, this is totally non-subjective so should fit in SO. Thank you.
Side note 1: I'll deal mostly with elongated, extended features like the top-right one, so circles are not that relevant.
Side note 2: To be 100% clear, I would need something (be it an already existant tool, or some ideas pointed out by you) that acts on the indices of the shapes, in terms of rows-columns into the original image, or on the boundary of the shape itself.
Side note 3: Apart from tools/suggestions/ideas, you are welcomed to write down some lines of code ;) I'm getting the regions as connected components from bwconncomp.
I had to solve a similar problem recently that involved counting the number of indentations on blobs within in an image (basically, the connected components returned by bwconncomp). The method I used was to look at curvature changes along the boundary calculated via the FFT. In your case, the red blobs would have a large number of curvature variations, whereas the black regions would not. It's a pretty easy calculation and relatively fast. The code is on github here:
https://github.com/mjsottile/blobdents
The file of interest is src/countindents.m. A short description of the approach is here:
http://arxiv.org/abs/1501.07692
I went for the easier road as suggested by #Mikhail in comments.
I found out regionprops has a really helpful tool called Solidity. Quoting docs,
Returns a scalar specifying the proportion of the pixels in the convex hull that are also in the region. Computed as Area/ConvexArea.
Convex hull is defined as the smallest convex polygon that can contain the region. So Solidity goes up to 1 if the shape is kind of regular and has no convexity changes; down to 0 for my red shape, which leaves space between itself and the convex polygon.
Of course it never reaches 0, lowest value should belong to a kind of +-shaped sign.
I have a CT scan for an heart and I am designing a device that rests on top of it. As such, getting the right lengths for certain attributes is important. The CT scan was segmented in MeshLab and my advisor gave me code that uses PLY_IO to read the ply file exported from MeshLab. From this, I have a map of the surface. surf(Map.X, Map.Y,Map.Z) outputs the 3D model. Now, what I would ideally want is to be able to select points graphically via the figure window and have Matlab either tell me what the points are or allow me to draw a geodesic line to determine its length. Question: Does anyone have any idea of how I could do this in a simple way?
Ultimately, just drawing on the figure might be ok too if I can just get it in the right orientation. Ideally, though, I would select the start and end point and then Matlab would graphically show a geodesic on the surface that I can later find the length of. I'm willing to do some programming for this, but hopefully there's something out there you guys might already know about.
One way to interactively extract points on a surface is to use datacursormode. Here's a simple example of how to get two points:
surf(peaks);
dcm_obj = datacursormode(gcf);
set(dcm_obj,'DisplayStyle','datatip',...
'SnapToDataVertex','off','Enable','on')
disp('Select first point then press any key')
pause
c_info{1} = getCursorInfo(dcm_obj);
disp('Select second point then press any key')
pause
c_info{2} = getCursorInfo(dcm_obj);
Note that if you (or the user) changes mode (e.g. by clicking the rotate button) in order to select the point, you will have to switch back to datacursor mode to move the datacursor again:
You should now have c_info{1}.position and c_info{2}.position which are two points on the surface. Calculating the geodesic is another matter - have a look on the File Exchange, see if there's anything around already that will do the job for the type of data you have already.
I am trying to specify a line across a blood vessel in an image, and then have matlab specify the edges of the vessel (which are contained within the line). The next part will be comparing changes in the distance between these edges over time (so across 1000x more images).
I have tried the following code to get started:
I = imread('Obj1.tif');
imshow(I,[]);
improfile
And I was looking at available methods to detect the edges from intensity along that line that gets plotted (tangents, maxima/minima etc) but I am not convinced this is the best method. I looked into other tools on matlab such as the Canny Method, Sobel etc, but the examples for all of these only show how to detect edges throughout the entire image. My coding skills are not sufficient to have the algorithms specified along a single line of the users choosing. Methods that I have looked at on pubmed also seem more complicated then perhaps I need.
Does anybody have any ideas or suggestions from the point that I am currently at?
Thank you
I am able to draw shapes using the UIBezierPath object. Now I want to identify different shapes drawn using this eg. Rectangle , Square , Triangle , Circle etc. Then next thing I want to do is that user should be able to select a particular shape and should be able to move the whole shape to different location on the screen. The actual requirement is even more complex , but If I could make this much then I can work out on the rest.
Any suggestion or links or points on how do I start with this is welcome . I am thinking of writing a separate view to handle every shape but not getting how do I do that..
Thank You all in advance !!
I recommend David Gelphman’s Programming with Quartz.
In his chapter “Drawing with Paths” he has a section on “Path Construction Primitives” which provides a crossroads:
If you use CGContextAddLineToPoint your user could make straight lines defined by known Cartesian points. You would use basic math to deduce the geometric shapes defined by those points.
If you use CGContextAddCurveToPoint your user could make curved lines defined by known points, and I’m pretty sure that those lines would run through the points, so you could still use basic math to determine at least an approximation of the types of shapes formed.
But if you use CGContextAddQuadCurveToPoint, the points define a framework outside of the drawn curve. You’d need more advanced math to determine the shapes formed by curves along tangents.
Gelphman also discusses “Path Utility Functions,” like getting a bounding box and checking whether a given point is inside the path.
As for moving the completed paths, I think you would use CGContextTranslateCTM.