Discover MakerZone

MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

Learn more

Discover what MATLAB® can do for your career.

Opportunities for recent engineering grads.

Apply Today

how to read a .txt file in matlab ?

Asked by preet on 4 Jul 2013

i want to read a file name is imData.txt which has 9600 arbitrary float numbers. and store in array of size 80*120 after read i want show image of that array??


Walter Roberson on 5 Jul 2013

Is there exactly one float value per row? Are there exactly 80 float value per row? Exactly 120? Is everything in one large column?

Is the second value in the file to become matrix location (2,1) or matrix location (1,2) ?

Are the columns delimited by spaces? By tabs? Is there a fixed width for each column and the columns can "run together" because of that?

What are the minimum and maximum values in the file?

preet on 6 Jul 2013

walter there 120 float values in a row.and total 80 rows. 2nd location is (1,2).

every value delimited by space dear. min value is 0.000000 and max is 619.420000

preet on 6 Jul 2013

@walter , values are in double or say float . but image in MATLAB will shown after conversion in uint8 wt should i do?




No products are associated with this question.

3 Answers

Answer by Walter Roberson on 6 Jul 2013
Accepted answer

For data in that format, load() should work, as would textscan(). You would not need to reshape() the data after either of those.

image() and imagesc() do not convert to uint8. However, when image() is passed floating point data (instead of uint8 or uint16) then image will expect the data to be in the range of 0 to 1, whereas imagesc() would rescale the floating point values so that the minimum mapped to 0 and the maximum mapped to 1.

What you should do about the fact your data ranges from 0 to 619 depends on your purposes. Are you okay with the loss of resolution if you rescale to 0 to 255? If so then

scaleddata = uint8(YourData ./ max(YourData(:)) .* 255);

Or you could leave the data in floating point format if you do not specifically need it to be uint8()

scaleddata = YourData ./ max(YourData(:);

The above assumes that for each file, the maximum value of the file should be treated as being full brightness. If there is some other value that should be treated as maximum brightness (e.g., 640) then use that value in place of max(YourData(:))


Walter Roberson
Answer by Sivakumaran Chandrasekaran on 4 Jul 2013

use textread command


preet on 5 Jul 2013

please explain it....i face problem during implemetation

Jan Simon on 5 Jul 2013

@kp: Then please show what you have tried and explain the problem. Currently too much guessing is required to give an answer.

preet on 6 Jul 2013

@ jan please read my previous comment after walter's comment, for explanation.

Sivakumaran Chandrasekaran
Answer by Hugo on 6 Jul 2013

Suppose that your file is called data.txt You can load the data simply with


which will create a variable called "data"

and the print it using



preet on 6 Jul 2013

may be its like




i think its like that

preet on 6 Jul 2013

but my values are in double but images will be shown after convrt in uint8 so wts the solution?

Hugo on 6 Jul 2013

If I understood right, you said that the values are in matrix form in the file. That should mean that your file is structures as lines ending in a newline symbol, and each number in each line is separated by a space. If that's the case, no matter whether the numbers are integer or real, load(...) will give you a matrix with as many rows as lines in the file and as many columns as numbers in the line, provided the number of numbers is the same in each line.

Once this is solve, I don't know what your values represent. If the position in the matrix represents the position of the pixels and the value represents the colour, then image(...) should show the image. Otherwise, I think it's up to you what you really have and how you really want to represent it.

I hope this helps you. Good luck.


Contact us