MATLAB: How to read .bin file

.binfilesMATLABread

I would like to read a .bin file in matlab

Note that all values are 16-bit signed integers stored in binary format (2 bytes – not human readable ASCII text).

Can anyone help me???

Best Answer

fid = fopen('YourFile.bin', 'r');
data = fread(fid, '*int16');
fclose(fid);

You might need to reshape() the data afterwards.

Depending on how the data was produced and what machine you are using to read it, you might find it necessary to swap the bytes in each integer. There is a MATLAB swapbytes() routine, or there is a FEX contribution http://www.mathworks.com/matlabcentral/fileexchange/17661-swapbytes-c-mex-function that is likely to be faster.

Alternately you might want to add the 'ieee-be' or 'ieee-le' option to the fread() call; that should save you from having to swap the bytes yourself, when the routine that produced the .bin file bothered to document the byte order.

Note: in general, .bin files are not 16 bit signed integers. ".bin" is any arbitrarily complicated binary format, and two .bin files produced by the same program will not necessarily be the same internal format.

Related Solutions

MATLAB: Fwrite and fread data of multiple precision to binary file

You need to fseek() to skip the first 6 bytes of the file.

MATLAB: How to open a part of data in the bin file

Unfortunately you wrote to the file instead of reading it, and you also completely clobbered the content of the file. I hope you had a backup copy.

You should have used something like,

fileID = fopen('name.bin', 'r');
Data = fread(fileID, [80 90], 'double');
fclose(fileID);

There are some pieces of information you need for binary files:

Is the file written as single precision floating point, or as double precision floating point, or as unsigned 8 bit integers, or as unsigned 16 bit integers, or as signed 16 bit integers? (There are other possibilities for integers, with both signed and unsigned 8, 16, 32, and 64 bit integers)
What is the "byte order" or "endianness" of the binary data. For example, if you had the 16 bit unsigned integer 258 stored in the file, then in hex that could be represented as 0x0102 -- 256^1 * 1 + 256^0 * 2 . That value could be written to the file either in the byte order [01 02] or in the byte order [02 01]. The order [01 02], where the "most significant" byte (the byte that contributes most to the overall value) occurs first in the file representation, is often called "big endian", and the order [02 01], where the "least significant" byte (the byte that contributes least to the overall value) occurs first in the file representation, is often called "little endian". and has a minor (mostly historical) advantage; see https://softwareengineering.stackexchange.com/questions/95556/what-is-the-advantage-of-little-endian-format/95854#95854 . All current systems that MATLAB is implemented on use "little endian" internally, so files written on x86 or x64 systems tend to be in "little endian" order. It is, though, an unnatural order for most cultures in the world, so the more "professional" order is "big endian".

These two considerations can alter the details of reading binary data.

Best Answer

Related Solutions

MATLAB: Fwrite and fread data of multiple precision to binary file

MATLAB: How to open a part of data in the bin file

Related Question