Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
convert binary to gray
1,466 views
Skip to first unread message
u-ingenieria bd
Jan 26, 2012, 1:53:22 PM1/26/12
to
how to convert a logical binary image into same size gray image?
I have tried y=bin2gray(x,'qam',16) function but here for x this
function does not take the binary image directly. all I need to
convert the binary image into grayscale image.
I have tried y=bin2gray(x,'qam',16) function but here for x this
function does not take the binary image directly. all I need to
convert the binary image into grayscale image.
ImageAnalyst
Jan 26, 2012, 3:31:47 PM1/26/12
to
On Jan 26, 10:53 am, u-ingenieria bd <u.ingenieria...@gmail.com>
wrote:
------------------------------------------------
I never heard of that function. Simply do this:
grayImage = uint8(binaryImage);
or, if you want the true (1) to be mapped to 255 instead of 1, do
this:
grayImage = 255 * uint8(binaryImage);
wrote:
I never heard of that function. Simply do this:
grayImage = uint8(binaryImage);
or, if you want the true (1) to be mapped to 255 instead of 1, do
this:
grayImage = 255 * uint8(binaryImage);
Steven_Lord
Jan 26, 2012, 4:39:58 PM1/26/12
to
"ImageAnalyst" <imagea...@mailinator.com> wrote in message
news:9a257a26-211e-4107...@iu7g2000pbc.googlegroups.com...
> On Jan 26, 10:53 am, u-ingenieria bd <u.ingenieria...@gmail.com>
> wrote:
>> how to convert a logical binary image into same size gray image?
>>
>> I have tried y=bin2gray(x,'qam',16) function but here for x this
>> function does not take the binary image directly. all I need to
>> convert the binary image into grayscale image.
>
> ------------------------------------------------
> I never heard of that function. Simply do this:
*snip*
> wrote:
>> how to convert a logical binary image into same size gray image?
>>
>> I have tried y=bin2gray(x,'qam',16) function but here for x this
>> function does not take the binary image directly. all I need to
>> convert the binary image into grayscale image.
>
> ------------------------------------------------
> I never heard of that function. Simply do this:
FYI, it's part of Communications System Toolbox:
http://www.mathworks.com/help/toolbox/comm/ref/bin2gray.html
and doesn't really appear, to my eyes, to be related to image processing.
--
Steve Lord
sl...@mathworks.com
To contact Technical Support use the Contact Us link on
http://www.mathworks.com
u-ingenieria bd
Jan 26, 2012, 8:41:40 PM1/26/12
to
On Jan 27, 3:39 am, "Steven_Lord" <sl...@mathworks.com> wrote:
> "ImageAnalyst" <imageanal...@mailinator.com> wrote in message
Ya the 255multiply is a must option to use when someone wants to see
the output with imshow() function. Its more interesting that if
someone wants to write that binary image into rgb format or png format
then he/she does not need to even convert the binary image into
grayscale or rgb. all it needs to use imwrite() function. Its
interesting.
> "ImageAnalyst" <imageanal...@mailinator.com> wrote in message
the output with imshow() function. Its more interesting that if
someone wants to write that binary image into rgb format or png format
then he/she does not need to even convert the binary image into
grayscale or rgb. all it needs to use imwrite() function. Its
interesting.
ImageAnalyst
Jan 26, 2012, 11:15:01 PM1/26/12
to
On Jan 26, 5:41 pm, u-ingenieria bd <u.ingenieria...@gmail.com> wrote:
> Ya the 255multiply is a must option to use when someone wants to see
> the output with imshow() function. Its more interesting that if
> someone wants to write that binary image into rgb format or png format
> then he/she does not need to even convert the binary image into
> grayscale or rgb. all it needs to use imwrite() function. Its
> interesting.
----------------------------------------------------------
> Ya the 255multiply is a must option to use when someone wants to see
> the output with imshow() function. Its more interesting that if
> someone wants to write that binary image into rgb format or png format
> then he/she does not need to even convert the binary image into
> grayscale or rgb. all it needs to use imwrite() function. Its
> interesting.
That's not correct. imshow will show a logical image automatically
scaled so that the true (1) pixels will appear bright (255). No
scaling is necessary to see it.
Yumnam Kirani
Jan 27, 2012, 3:25:10 AM1/27/12
to
ImageAnalyst <imagea...@mailinator.com> wrote in message <9a257a26-211e-4107...@iu7g2000pbc.googlegroups.com>...
It is not possible to convert a binary image to a gray image, irrespective of whether we use uint8 or multiply by 255 or both. The resulted values will only have two distinct values, that is, binary values.
Yumnam Kirani Singh
Tronglaobi Awang Leikai
Yumnam Kirani Singh
Tronglaobi Awang Leikai
Carl S.
Dec 27, 2012, 12:42:06 PM12/27/12
to
ImageAnalyst <imagea...@mailinator.com> wrote in message <9a257a26-211e-4107...@iu7g2000pbc.googlegroups.com>...
I have two images. They are;
tempBinaryImg : 320x320 double
GrayscaleImg : 320x320 double
When I write that
temp2 = tempBinaryImg.*(GrayscaleImg>0);
PreprocessedGrayscaleImg = uint8(Img).*uint8(temp2);
then I am getting PreprocessedGrayscaleImg as uint8. But I want to get the PreprocessedGrayscaleImg as double. So, I write
PgrayImg = double(PreprocessedGrayscaleImg);
but the PgrayImg is binary, it is not gray. How can I get the PreprocessedGrayscaleImg as double and gray ?
Image Analyst
Dec 27, 2012, 5:08:08 PM12/27/12
to
"Carl S." wrote in message <kbi19e$f3q$1...@newscl01ah.mathworks.com>...
It's not binary. It's a gray scale image of type "double". Why? Do "whos PgrayImg" to see the data type of PgrayImg, or else look in the Workspace panel. But why does it have values of only 0.0 and 1.0? Well let's take it a step at a time:
(GrayscaleImg>0) is a logical (0's and 1's), but gets converted to double to do the multiplication.
tempBinaryImg is double, but judging from the name, it also has values of 0's and 1's.
So the product temp2, will also be a double but have values of 0.0 and 1.0 only.
So now you cast temp2 to uint8 which means it's now unsigned 8 bit integer and will have integer (not floating point) values of 0 and 1.
Then you have some array "img" and I have no idea what values it has, but after you cast it to uint8, it will have integer values of 0, 1, 2, 3, ....255, or whatever it actually contains.
So you do uint8(Img).*uint8(temp2) and you will get PreprocessedGrayscaleImg as a uint8 also, and it will have values of 0, 2, 3, 4, ... 255 or the same values that img, masked by temp2, would have. And now you do this:
PgrayImg = double(PreprocessedGrayscaleImg);
which means that PgrayImg has the same values as PreprocessedGrayscaleImg but they are double instead of integer, in other words floating point 0.0, 1.0, 2.0, 3.0, ...255.0, or whatever values the masked version of tempBinaryImg has.
You said you think it's binary (type logical), but it's not - it's double, though it might have floating point values of 0.0 and 1.0 only, if that's what tempBinaryImg had.
> I have two images. They are;
> tempBinaryImg : 320x320 double
> GrayscaleImg : 320x320 double
>
> When I write that
> temp2 = tempBinaryImg.*(GrayscaleImg>0);
> PreprocessedGrayscaleImg = uint8(Img).*uint8(temp2);
>
> then I am getting PreprocessedGrayscaleImg as uint8. But I want to get the PreprocessedGrayscaleImg as double. So, I write
>
> PgrayImg = double(PreprocessedGrayscaleImg);
>
> but the PgrayImg is binary, it is not gray. How can I get the PreprocessedGrayscaleImg as double and gray ?
----------------------------------------------------------------------------------------------------------
> tempBinaryImg : 320x320 double
> GrayscaleImg : 320x320 double
>
> When I write that
> temp2 = tempBinaryImg.*(GrayscaleImg>0);
> PreprocessedGrayscaleImg = uint8(Img).*uint8(temp2);
>
> then I am getting PreprocessedGrayscaleImg as uint8. But I want to get the PreprocessedGrayscaleImg as double. So, I write
>
> PgrayImg = double(PreprocessedGrayscaleImg);
>
> but the PgrayImg is binary, it is not gray. How can I get the PreprocessedGrayscaleImg as double and gray ?
It's not binary. It's a gray scale image of type "double". Why? Do "whos PgrayImg" to see the data type of PgrayImg, or else look in the Workspace panel. But why does it have values of only 0.0 and 1.0? Well let's take it a step at a time:
(GrayscaleImg>0) is a logical (0's and 1's), but gets converted to double to do the multiplication.
tempBinaryImg is double, but judging from the name, it also has values of 0's and 1's.
So the product temp2, will also be a double but have values of 0.0 and 1.0 only.
So now you cast temp2 to uint8 which means it's now unsigned 8 bit integer and will have integer (not floating point) values of 0 and 1.
Then you have some array "img" and I have no idea what values it has, but after you cast it to uint8, it will have integer values of 0, 1, 2, 3, ....255, or whatever it actually contains.
So you do uint8(Img).*uint8(temp2) and you will get PreprocessedGrayscaleImg as a uint8 also, and it will have values of 0, 2, 3, 4, ... 255 or the same values that img, masked by temp2, would have. And now you do this:
PgrayImg = double(PreprocessedGrayscaleImg);
which means that PgrayImg has the same values as PreprocessedGrayscaleImg but they are double instead of integer, in other words floating point 0.0, 1.0, 2.0, 3.0, ...255.0, or whatever values the masked version of tempBinaryImg has.
You said you think it's binary (type logical), but it's not - it's double, though it might have floating point values of 0.0 and 1.0 only, if that's what tempBinaryImg had.
0 new messages