How 'imread' function works in Matlab?

Question

Mohammed Sadeq 2017-2-21

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/326169-how-imread-function-works-in-matlab

评论： Mohammed Sadeq 2017-2-22

采纳的回答： Walter Roberson

ImreadReturn.mat

Hi,

I am using 'imread' function in Matlab to read images produced by X-Ray CT. Those images are in the greyscale.

When I use the 'imread' function, Matlab returns an array with the same size of the image but with numbers inside the cells (matrix is attached). The number represent a single pixel in that image while the minimum number is 3107 and the maximum is 40899.

I want to know how those numbers are calculated?

Thanks..

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Walter Roberson 2017-2-22

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/326169-how-imread-function-works-in-matlab#answer_255838

The png file you posted is 8 bits per channel, uint8, 3 channel, with a transparency (alpha) channel that is set to all 255 (uint8) which is completely opaque.

The .mat file you posted relates to a different image, and is a matrix of uint16 values.

Are you asking how the uint16 is represented inside the file? If so then the answer is that they are stored in http://stackoverflow.com/questions/2384111/png-file-format-endianness "Network Byte Order", more commonly known as "Big Endian". That is a byte order in which the most significant of the bytes occurs first and then the less significant. That is the opposite of the order used by the x86 architecture, but it is the same as the order reflected by dec2hex and sprintf() with '%x' format.

What do the values correspond to, what units are they? I would put it to you that the images have been extracted from DICOM files without the relevant headers. See http://www.idlcoyote.com/fileio_tips/hounsfield.html

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Mohammed Sadeq 2017-2-22

Thanks for your complete answer. I will accept it.

请先登录，再进行评论。

Answer 2

Adam 2017-2-21

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/326169-how-imread-function-works-in-matlab#answer_255717

They aren't 'calculated', they are read in from the file as the name 'imread' and the help suggest.

2 个评论
显示无隐藏无

Mohammed Sadeq 2017-2-22

Hi Adam,

If they are read from the image it means that they are calculated (or measured) somehow based on bits in each pixel. The numbers (3107 and 40899) came from somewhere. I want to know from where.

Thanks for your answer..

Adam 2017-2-22

imread literally reads what is stored in the file. How those numbers were calculated comes from whatever process was used to create the image and save it originally.

请先登录，再进行评论。