convert an 8 bit image into lower bits

Question

Ghulam 2017-12-5

1
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/371120-convert-an-8-bit-image-into-lower-bits

评论： Walter Roberson 2023-2-9

Hi, I want to convert an 8bit gray scale image into lower bits. I have loaded the image and divided it by multiple numbers to downsize its number of bits, but the problem is that the result is not presenting the true form of White color, and the white color is removed with downsizing. Can you please tell me what is the correct method of downsizing the bits of an image. Regards

8 个评论
显示 6更早的评论隐藏 6更早的评论

Walter Roberson 2023-2-9

在 MATLAB Online 中打开

@Alicia Davion

A 1 bit image cannot represent three different states. At any one location, a 1 bit image can have only 2^1 different states, 0 or 1. A 2 bit image can have 2^2 different states, binary 00 01 10 11 which are equivalent to decimal 0 1 2 3 . A 3 bit image can have 2^3 different states, binary 000 001 010 011 100 101 110 111 which are equivalent to decimal 0 1 2 3 4 5 6 7. This is the same way that a single-digit decimal number can have only 10^1 states, zero through nine, and a two-digit decimal number can have only 10^2 states, zero through ninety-nine.

If you need to represent 0 1 and 2 then you need a minimum of 2 bits per location. Though if you pack several locations together you can do better:

capacity3 = floor(log(2^32)/log(3))
capacity3 = 20

so 32 bits can represent twenty base-3 digits instead of 32/2 = 16 digits if you did not pack the information densely.

If you are working with integer classes watch out for the fact that integer arithmetic rounds

IN = uint8(0:32:255)
IN = 1×8
     0    32    64    96   128   160   192   224
IN/128
ans = 1×8
   0   0   1   1   1   1   2   2

To prevent that you need to do something like

uint8(floor(double(IN)/128))
ans = 1×8
   0   0   0   0   1   1   1   1

Steven Lord 2023-2-9

在 MATLAB Online 中打开

If you use the idivide function instead of / you can control how the result is rounded.

IN = uint8(0:32:255);
result = [IN; IN/128;
idivide(IN, 128, 'fix');
idivide(IN, 128, 'floor');
idivide(IN, 128, 'ceil');
idivide(IN, 128, 'round')]
result = 6×8
     0    32    64    96   128   160   192   224
     0     0     1     1     1     1     2     2
     0     0     0     0     1     1     1     1
     0     0     0     0     1     1     1     1
     0     1     1     1     1     2     2     2
     0     0     1     1     1     1     2     2

Walter Roberson 2023-2-9

Thanks, @Steven Lord I didn't know about idivide !

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Walter Roberson 2017-12-5

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/371120-convert-an-8-bit-image-into-lower-bits#answer_294738

在 MATLAB Online 中打开

img8 = imread('cameraman.tif');
im4 = im8 ./ 16;
image(im4);
colormap(gray(16));

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。

convert an 8 bit image into lower bits

8 个评论
显示 6更早的评论隐藏 6更早的评论

采纳的回答

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

更多回答（0 个）

另请参阅

类别

标签

Community Treasure Hunt

convert an 8 bit image into lower bits

8 个评论 显示 6更早的评论隐藏 6更早的评论

采纳的回答

0 个评论 显示 -2更早的评论隐藏 -2更早的评论

更多回答（0 个）

另请参阅

类别

标签

Community Treasure Hunt

WeChat

8 个评论
显示 6更早的评论隐藏 6更早的评论

0 个评论
显示 -2更早的评论隐藏 -2更早的评论