General problem description
I have 33 TIFF16 images and I want to do some processing on them using MATLAB. So reading them is the first step. After I download an image from the web and then try to read it using MATLAB's imread (as well as Tiff and read). I display the image using imshow. The image displayed by the Windows File Viewer and MATLAB is totally different. I cannot process them since I don't trust MATLAB has read them correctly. I give more specifics of the problem now.
EDIT: If it helps, the details of the TIFF16 images are: TIFF (16 bits per channel, ProPhoto RGB color space, lossless compression)
More details:
I download an image a0008-WP_CRW_3959.tif. Destination: Go to this link -> img0008 -> Expert B (In case somebody wants to try, otherwise I have screenshots below).
I read that image in MATLAB using: img=imread('imgFilename.tif','tiff'); imshow(img,[]); or
t = Tiff('imgFilename.tif','r');
imageData = read(t);
imshow(imageData);
Now, I display the snapshots of Windows file viewer:
Next, snapshot of what MATLAB shows me:
Now, I have a good reason to believe that Windows file viewer is correct. Go the same link as previous. Scroll down to img0008. Hover your mouse onto the leftmost img0008. A thumbnail view of Expert B will come which looks same as what Windows shows me.
Does anybody know how to make MATLAB read and show the tiff16 image correctly?
Thank you #MarkRansom for pointing me to the embedded color profile possibility. I believe the following solution is correct and produces the same output as Windows File Viewer.
First read the icc-color-profile using iccread command.
I_rgb = imread('a0008-WP_CRW_3959.tif');
outprof = iccread('sRGB.icm');
P = iccread('a0008-WP_CRW_3959.tif');
Then convert the image into sRGB profile using makecform and applycform:
C = makecform('icc',P,outprof);
I_cmyk = applycform(I_rgb,C);
imwrite(I_cmyk,'pep_cmyk.tif','tif')
info = imfinfo('pep_cmyk.tif');
imshow('pep_cmyk.tif');
The original image saved on disk and the new - pep_cmyk.tif - look exactly the same with Windows file viewer.
Related
I'm working with fastai, trying to pass some images to a dataloader. The original images are kind of pinkish, but after passing them they appear mostly as green-black (see image in link below):
Original pinkish image (up) and example images (down) after passing them to dataloader, and the code.
The code I've used for the datablock and to show the images is:
example = DataBlock(
blocks=(ImageBlock, CategoryBlock),
get_items=get_image_files,
splitter=GrandparentSplitter(),
get_y=parent_label,
item_tfms=Resize(128)) #already tried it without item_tfms just in case, still black-green
dls = example.dataloaders(path)
dls.show_batch(nrows=1, ncols=3)
I tried with .tif and .jpeg images, and both show the same problem. The only thing that comes to my mind is that somewhat somewhere is not reading correctly the color format (RGB according to my original files), or maybe transforming it; but I'm not able to figure it out.
Just in case it's important, I'm working in a Jupyter notebook with a MackBook Air M1.
Thanks!
Irene
HELLLPPPPP!!!!! I am programming in Corona SDK, and I am trying to insert a backround image, with the following code ---local backround = display.newImage("bluebackround.jpg")--- But it does not want to show up in the simulator. It keeps giving me the following message in the output, ---Failed to find image 'bluebackround.jpg'---. My image is saved in the project folder. I am using Microsoft Windows software. I have done another project using Corona, and I inserted images completely fine. Anyone know what's going on with my code and how to fix it? Thank you very much in advance.
_W = display.contentWidth;
_H = display.contentHeight;
local image = display.newImageRect(--[["Name", width, heigth]] "bluebackround.jpg",_W, _H)
image.x = _W/2;
image.y = _H/2;
The your image need to be in the same folder as main.lua in your case. On the same level. Moreover, I think you misspell name of your image. You write 'bluebackround' without letter g in middle. So check name of the file.
From Corona documentation
Image guildelines:
Corona supports PNG and JPG format,
Images should not contain an embedded ICC profile,
Avoid progressive JPG files since they will take much longer to load.
I want to make screenshots on OSX using python. I dont want make full screen shots but only certain rectangles on the screen. Something like (291,305,213,31). I need the correct pixel because afterwards the image files are processed by OCR (python-tesseract) to extract the text.
By the way this is since 6 years the first time I am programming, so far I only know Java a bit. I started yesterday and gave up this morning at 4am. So basically I have no clue yet...For example I still cannot build with Sublime because of path settings, but thats a different story. Cant figure out everything on one day.
I was trying already the following:
- wxPython
But the result are black images, see also:
stackoverflow.com/questions/8644908/take-screenshot-in-python-cross-platform
Additionally it only works in 32-bit mode, but when I do OCR using python-tesseract openCV requires 64-bit....
autopy
when trying to install I got errors, see also:
stackoverflow.com/questions/12993126/errors-while-installing-python-autopy
ImageGrab
only Windows
effbot.org/imagingbook/imagegrab.htm
commandline screencapture
os.system('screencapture test.png')
When I found this I thought, nice but only fullscreen when checking man screencapture. But then I found this: guides.macrumors.c om/screencapture
-R capture screen rect
That would be already enough, but on OSX 10.7.5 I dont have this option. Any ideas?
import Quartz.CoreGraphics
neverfear.org/blog/view/156/OS_X_Screen_capture_from_Python_PyObjC
Create screenshot as CGImage
image = CG.CGWindowListCreateImage(
region,
CG.kCGWindowListOptionOnScreenOnly,
CG.kCGNullWindowID,
CG.kCGWindowImageDefault)
Unfortunately the image is not in file format but a CGImage, no idea how to save as file.
So if possible I would like to use the commandline screencapture with -R if somebody knows how. Just as a start to continue.
Are there any other command line tools available?
What about other libs that I have missed?
Cheers
M
Given that you can get a CGImageRef, you can get its pixel data using the techniques described in Technical Q&A QA1509: Getting the pixel data from a CGImage object. In particular, it shows a function to get the pixel data as a CFDataRef using this function:
CFDataRef CopyImagePixels(CGImageRef inImage) { return CGDataProviderCopyData(CGImageGetDataProvider(inImage)); }
and says:
The pixel data returned by CGDataProviderCopyData has not been color
matched and is in the format that the image is in, as described by the
various CGImageGet functions …
It shows an alternative for getting the pixel data in other formats if you need that.
Basically, I'd like to resize or resample a .png image (in order to reduce its file size) and yet retain it's transparency.
Anybody got an idea how best to go about this?
Thanks.
You can use paint.net, it is a free tool. Although it is pretty basic, it does the job.
Go to Image > Resize
Stumbled upon this thread and found the following site that does exactly what is requested: https://onlinepngtools.com/resize-png
What graphics program are you using?
Photoshop does this by simply going thru IMAGE > IMAGE SIZE and resizing. Transparency is not affected.
I'm sure Paint Shop Pro does the same
I know this is an old question, but the answer that worked for me was to use Inkscape.
Start Inkscape (free on Inkscape.org).
File -> Import... (Ctrl+I) the PNG file you want to resize (defaults on import dialog are ok).
With the image you just imported selected, select File -> Export PNG image... (Shift+Ctrl+E)
In the Export PNG Image tool pane, click the Export As... button to set the output filename and location.
In this same tool pane, set the image size using width/height or pixels.
In this same tool pane, click the Export button to create the output file.
This worked for me, hope it helps someone else.
Providing the image you have created / have been working on is transparent in the first place, using the "Resize" or "Resample" tools in any major image editing package (e.g. PhotoShop, PaintShop Pro and so on) should not affect (or lose) the transparency at all.
I use PaintShop Pro (X6, 64 Bit) myself and typically find that the "PNG Optimizer" option offers more options along these lines (than the default "Save As > .png" route).
Hope that this helps (specific to PaintShop Pro Users) in relation to the source question.
While I was waiting for the downloads of other image editing softwares, I tried Microsoft Power Point and succeeded in preserving the transparency.
Drag the image inside any slide, crop or resize, then save as a new picture as .png.
You can drag the image back in ppt to confirm the transparency is maintained
The complete Autodesk Sketchbook is now out for free including all the previous premium features such as resizing an image.
You do it as you would in paint by clicking on Image > Image Size... and then you can save as a .png without losing transparency.
Image size can be reduced by reducing number of colors and there are online tools to do this .
Try these..Hope they solve your problem
https://tinypng.com/
http://pngcrush.com/ and
http://tools.dynamicdrive.com/imageoptimizer/ --It provides more output images with different number of colors. However, smoothness will be effected, take care.
I am a Mac (10.6.8) user. I have written MATLAB code for plotting calculation results and then saving the plots as pdfs. I use "saveas" (see example below).
I get this error:
??? Error using ==> print at 325
Problem converting PostScript. System returned error: -1.Failed to convert to output format; Ghostscript status: -100.**** Unable to open the initial device, quitting.
Error in ==> saveas at 155
print( h, name, ['-d' dev{i}] )
Error in ==> Results_processor at 1219
saveas(gcf,saveFigTo1, 'pdf')
Here is the relevant piece of code:
calculationResultsPath = '/Me/Projects/ThisProject';
calculationResultsDirectory = strcat( calculationResultsPath,'MATLABProcessedResults' );
mkdir( calculationResultsDirectory );
% ...Code for importing results to be plotted from external files (works fine)...
% ...Code for plotting (works fine)... I get the figures I want.
% The problem is:
saveFigTo1 = strcat(resultsDirectory,'/majorsMgO.pdf')
saveas(gcf,saveFigTo1, 'pdf')
hold off
pause
clf;
Some further information... This worked fine last week when I first wrote it! Since then, I think I updated to Mac OS 10.6.8 from 10.6.7, but nothing else has changed in my code or in the version of Matlab I use (R2009a) (unless I have a very bad memory!).
Also, I came across some older suggestions on similar questions to use "print." I tried using:
print(gcf,'filename'). I did get a pdf, but it would not open in any pdf viewing programme. I assume (but don't know for sure) that this could be related to the fact that I use a Mac. I have noticed that there are a few things (especially related to external file manipulation) that don't run on Mac.
If anyone could help, I would be very grateful.
Update:
I found a GhostScript for Mac and installed it as Chris suggested. Unfortunately, that didn't work. I read on a forum that many Mac users are having problems with MATLAB plots at present, possibly related to java. There was an operating system update (to OS X 10.6.8) last week and that is when the problems began. My code worked before then.
I still have not found a solution to this and I don't think the MATLAB people have either, so if anyone has suggestions about how to save plots WITHOUT using saveas, I would love to hear them. The "print" command does not work for me either--it produces PDFs that I can't open.
I think the problem here is GhostScript dying and not so much Matlab. A Google for that GS error turns up many pages such as this. Is this at all applicable? If you use GS outside of Matlab does it work?
As an aside, you might take a look at this FEX submission export_fig. It has treated me well. Worst case you could output to png and convert to PDF later.
One recommendation I can make is to use a different format that OS X understands and simply convert the result to pdf with a system call
See if the following works:
% Load a test image
im = imread('cameraman.tif');
imshow(im); % display the image
saveas(gcf,'test.tif','tif');
% convert to pdf using a syscall to cupsfilter
!cupsfilter test.tif > test.pdf 2> /dev/null
% open the file with your default pdf viewer
!open test.pdf
If the above is not working, yet another approach is to get the bitmap from the figure window and write it using imwrite. Note, this method doesn't benefit from the nice font scaling features of the saveas and print. This approach assumes the im variable above still exists.
imagesc(im); colormap gray;
% Set the border color to white
set(h,'Color',[1 1 1]);
% Get the image in the figure
frame = getframe(gcf);
imout = frame.cdata;
% on OS X, the stretch window image
% appears in the bottom right corner
% of the image. Remove it.
imout = imout(10:end-9,10:end-9,:);
% Write the image out to a lossless tif
imwrite(imout,'test.tif','tif','Compression','none')
You could then convert the tifs to pdf files as above. The quality of the figure will depend on the figure size. In most cases, I wouldn't use this second approach as saveas handles the fonts nicely. Using getframe is meant only as a workaround to the true problem with saveas.