MATLAB: How to extract numeric data from a textfile

coordinatedataplotscantext filexy

Hi all, is there a method in which I would be able to extract x and y coordinate data from a textfile, when the x and y values are divided by a '/'?

I need to extract and then plot the coordinate data.

The textfile should be attached above. Any help would be greatly appreciated. Thanks 🙂

Best Answer

S = fileread('P01-GroupM(1).txt');
XYstrs = regexp(S, '^X/Ycoords \(in):\s*(?<X>\S+)\s*/\s*(?<Y>\S+)', 'names', 'lineanchors', 'dotexceptnewline');
X = str2double({XYstrs.X});
Y = str2double({XYstrs.Y});
plot(X, Y)
axis equal

Related Solutions

MATLAB: How to replace a portion of data with another

This works. The example input and output files are attached.

newData = [1,2;3,4;5,6;7,8];
newName = 'blah.prw';
fid = fopen('oldfile.txt', 'rt');
fod = fopen('newfile.txt', 'wt');
for k = 1:3
    fprintf(fod,'%s\n',fgetl(fid));
end
str = regexprep(fgetl(fid),'\w+\.prw',newName);
fprintf(fod,'%s\n',str)
cmp = ' !-';
while ~isequal(intersect(str,cmp),cmp)
    str = fgetl(fid);
    fprintf(fod,'%s\n',str)
end
fprintf(fod,'          %.7f          %.7f\n',newData.')
while ~isempty(str)
    str = strtrim(fgetl(fid))
end
fprintf(fod,'          \n')
while ~feof(fid)
    fprintf(fod,'%s\n',fgetl(fid))
end
fclose(fid);
fclose(fod);

MATLAB: How to show an image from a textfile

jpg images are typically colour images. Colour images are stored as 3D arrays in matlab. dlmread and dlmwrite can only read 2D arrays so your serialisation to text loses that third dimension. All three pages are concatenated horizontally. In addition, jpg images are typically of type uint8, these values get cast to double without rescaling when you serialise them.

You can get back your original array by reshaping it back to 3D and casting it back to uint8:

readimage = dlmread('FileName23.txt');
readimage = uint8(reshape(readimage, size(readimage, 1), [], 3));
imshow(readimage);

Note that this applies to 8-bit colour jpg, which is the majority of jpg files. If the original jpg was 16-bit then you'd have to cast to uint16. If the original jpg was greyscale, you don't need the reshape.

Best Answer

Related Solutions

MATLAB: How to replace a portion of data with another

MATLAB: How to show an image from a textfile

Related Question