I am trying to port some old matlab code to python. I chose OpenCV as I am familiar with the library. Despite that, I found results differ a bit (this program seems very sensible to small changes in texture), and I found pixel values are sightly different even with just reading the image from disk (I thought It could be some antialiasing or odd behavior when reecaling, but its there even before modifying anything)
I am aware of the different color order (RGB in matlab by default, BGR on OpenCV), but still pixel values are sometimes off by +-2 units (on 8-bit per color images). See for example in the following screencap, second pixel is 5-14-9 (RGB) when in matlab its 5-14-11. First pixel is exactly the same value.
I can't think of any way to check the EXACT transformation/rounding that matlab is performing, or why this is different in the first place. Any Ideas on this matter?
Are you sure that you are looking at the correct pixel?
Matlab and Python differ in indexing, in Matlab the first index is 1, and in Python the first index is 0.
My guess is that you should be comparing the Matlab pixel [2,1] with the Python pixel value at index 0, which is 5-14-11 like the one in Matlab.
Related
I am working on detection of dental issues using thermography. I need to separate the all the given colours in the image into separate clusters (4-7 in number) so that the high-temperature zones (seen white in the image) are seen separately, which can be followed by thresholding if need be.
I am also attaching a sample of the images I will be working on. I am looking for a suitable program to carry out the execution in MATLAB.
I've already worked on the same, the program being attached in the previous question, which gives 3 clusters only.Since I'm a beginner, I need help in establishing more clusters.
image obtained using thermal camera on which clustering is to be carried out
the closest I could get to the sort of clustering I want to carry out. here, in this image green-blue cluster and white cluster are in the same image, which i want to have in separate clusters, hence the need of more clusters
expected result after clustering and thresholding
Rather than hoping that by chance clustering does what you need, I'd rather just use the ground truth you have...
In case you haven't noticed: there is a color index to the right. That is an easy to use, ordered (this is extremely beneficial over clustering, in particular to set thresholds) easy to detect key to interpreting these images without hoping on a random generator.
Note that you will likely also need to read off the numbers that give the color scale, in order to compare images.
You can read your image file in Matlab and then convert the data from RGB format to HSL or HSV with the function rgb2hsl() or rgb2hsv().
They are two alternative representations of the RGB color model. Then you can easily make your discrimination with the value of H which is abbreviated for hue.
For more information take a look at the following link: HLS and HSV
I have a .png image that has been created from some grayscale numbers using Matlab's imagesc tool using the standard color map.
For some reason, I am unable to recover the raw data. Is there a way of recovering the raw data from the image? I tried rgb2gray which more or less worked, but if I replug the new image into imagesc, it gives me a slightly different result. Also, the pixel with the most intensity differs in both images.
So, to clarify: I would love to know, how Matlab applies the rgb colormap to the grayscale values, when using the standard colormap.
This is the image we are talking about:
http://imgur.com/qFsGrWw.png
Thank you!
No, you will not get the right data if you are using the standard colormap, or jet.
Generally, its a very bad thing to try to reverse engineer plots, as they will never contain the entirety of the information. This is true in general, but even more if you use colormaps that are do not change accordingly with the data. The amount of blue in jet is massively bigger in range than the amount of orange, or another color. The color changes are non-linear with the data changes, and this will make you miss a lot of resolution. You may know what value orange corresponds to, but blue will be a very wide range of possible values.
In short:
Triying to get data from representation of data (i.e. plots) is a terrible idea
jet is a terrible idea
Let me explain.
My program takes an x-ray in a format of the x-ray detector ".his" which goes from 0 to 65535, and from those values it can tell you how much of a certain material is in each pixel "4 cm of aluminum" for example.
It does that for every pixel and you end up with a matrix that tells you how much of a given material is there, and you can visualize that matrix and see only fat tissue in an image without the bones blocking your view, it's very cool I know.
What I want to do now is to save that matrix as an image so that I can analyse and modify that image with programs like Image J, but I also want that if I see the pixel value I see the original value, I want to see "4" and know that pixel shows 4 cm of lungs or whatever material I'm working on.
Is that possible?, my professor seems to think that it is but he's not sure how to do it, and figure that out is my job anyway.
It should be possible since with Image J I can open the ".his" format and I can do just that, I can see the values from 0 to 65535, provided I say Image J that the image is 16 bit unsigned and other properties of that kind of files, but I wouldn't know how to do that for a Matlab variable.
Thanks a lot.
So if I understand correctly, you want to save an image that also contains arbitrary metadata on every pixel (in this case an integer).
If you use an image format like PNG you could encode that extra data into the alpha channel (which would be nearly imperceptible with a value like 4/255 away from fully opaque), but you'd have to be careful when editing the image that you don't change the alpha channel by mistake.
However, this is rather finnicky and would be cumbersome to implement in Matlab.
Instead I would suggest simply saving a standard image and a text file (or binary file) alongside it with the data you want.
I have a 16bit voxel data set from which I need to extract the integer values for each voxel. The data set can be downloaded from here, it is the 'Head Aneuyrism 16Bits' data set (You need to click on the blood vessels image to download the 16bit version). Its size is 512x512x512, but I don't know whether it is greyscale or color, nor if that matters. Looking at the image on the website I'd guess that it is color, but I am not sure whether the image should be taken literally.
A related question on SO is the following: How can I read in a RAW image in MATLAB?
and the following on mathworks: http://www.mathworks.com/matlabcentral/answers/63311-how-to-read-an-n-dimensioned-matrix-from-a-binary-file
Thanks to the information in the answers to these questions I managed to extract some information from the file with matlab as follows:
fileID=fopen('vertebra16.raw','r');
A=fread(fileID,512*512*512,'int16');
B=reshape(A,[512 512 512]);
I don't need to visualise the image, I only need to have the integer values for each voxel, but I am not sure whether I am reading the information in the correct way with my script.
The only way I found to try and check whether I have the correct voxel values is to visualise B using the following:
implay(B)
Now, with the code above, and then using implay(B) I get a black and white movie with a white disc in the center and black background and some black pixels moving in the disc (I tried to upload a frame of the movie, but it didn't work). Looking at the image on the website from which I downloaded the file, the movie frames I get seem quite different from that image, so I'd conclude that I do not have the correct voxel values.
Here are some questions related to my problem:
Do I need to know whether the image is in grey scale or color to read the voxel values correctly?
On the data set website there is only written that the data set is in 16bit format, so how do I know whether I am dealing with signed or unsigned integers?
In the SO question linked to above they use 'uint8=>uint8'. I could not find this in the matlab manual, so I wonder whether 'uint8=>uint8' is an obsolete matlab notation for 'uint8' or if it does something different. I suspect that it does something different since if I use 'int16=>int16' instead of 'int16' in my code above I get a completely black movie with implay.
It looks like you read the data correctly.
The problem when displaying it is the scale of the values. implay seems to assume the values to be in [0,1] and therefore clamps all values to be in that range, where are your data range is [0,3000].
Simply doing
B = B / max(B(:))
will rescale your data to [0,1] and looking at the data again with
implay(B)
shows you something much more sensible.
The info for DensityPlot says that the "default generates colorized grayscale output, in which larger values are shown lighter." What on earth is colorized grayscale? Is there a way to make it truly grayscale without the blue and purple colors that it generates? And when I do it, it appears a little pixellated. Is there a way to evaluate it at more points so that it doesnt look so choppy?
By colorized grayscale, I think they mean that it's monochrome, or maybe bichromatic - that is, there's a linear scale from one color to another, rather than fully varying across the whole color space. It's not a very good term, I agree.
Specifying ColorFunction->GrayLevel should give you pure grayscale. This is distinct from the built-in gradient GrayTones (ColorFunction->"GrayTones"), which appears to stop a bit short of pure black and white on the ends and is a bit warm. There are plenty of other built-in gradients - see the return value of ColorData["Gradients"]. You can also specify your own function, of course - it will take as input a real number from 0 to 1, and should return a color specification, e.g. the return values of GrayLevel, RGBColor, Hue, or CMYKColor.
To make it less choppy, as with basically all plotting functions, try specifying a higher value for PlotPoints (the number of initial sampling points) or MaxRecursion (how many times it can resample).