The last time I purchased a piece of software to edit photographs, it was Photoshop 6 for PC. I had tried version 7, which did the job well, on others’ computers, but version 6 did the job, and did it for years on end and many computers.
However, I can’t install it anymore. Perfectly normal for software that came out ten years ago.
I have tried pretty hard to get GIMP to do what I want to do, but I don’t know how. I don’t want to invest in a version of Photoshop nowadays and find out that it doesn’t either, and that my hard-won reflexes are no longer current anyhow, which is the only reason I would upgrade.
Anyone know of a resource that would tell me how to do the following in either a modern version of PS that I could buy, or in GIMP?
This is sort of my workflow when I put together figures for papers:
- I open up all my photos that I want in the final figure. Usually TIFF but sometimes JPEG format
- Design a transparent base document with the final pixels-per-inch (or more) resolution
- Drag guidelines over from the rulers and place them where I want to frame each part of the figure (they’re usually multipart, lettered a, b, c etc.) and leaving an equal measurement of white space around them
- Note that these parts are not necessarily a reproducible size. They will take into account the shape of the object I want to show, so may be squares or rectangles.
- Paste in each photograph as its own layer, onto that base document
- Arrange them in position by dragging them around
- Perform a homotethic transformation – I don’t know what it is in normal English – but fit the part of the photo I want by enlarging or shrinking, to scale, the original photograph so that what I want to showcase appears in the guidelines I consecrate to that particular subfigure part. Know what the scale factor was, because most of my photos do not have automatic scale bars written in and I often have to calculate them and add later.
- Be able to rotate the photos at any angle so as to best adapt my axes to the frame.
- Be able to rename the layers by the letter of the part of the figure to which it corresponds.
- Adjust contrast and color so that the backgrounds of comparable photographs are similar, if they were not all taken at once under identical conditions (usually the case)
- Be able to clone-stamp out bits of dust in the background or if they are in a non-essential edge of the photo. I believe that much clean-up is permitted, and I’ve always done that out of respect for the visual message.
- Be able to annotate with Arial text in any color, as a separate layer (or more).
- Be able to add any color curved or straight lines, solid or dashed, and arrows and arrowheads, as a separate layer (or more). Be able to duplicate these additional forms.
- Save the entire figure with its layers.
- Save the entire figure with all layers fused into one TIFF file.
Suggestions, please! Links also welcome.
Heather, I have used many graphics packages for a long time, including PS7-CS3, GIMP and the NIH ImageJ. I can say with 100% certainty that nothing beats Photoshop. All the parts of the workflow you mentioned can be done in Photoshop, and therefore, it will be a great investment if you buy it. You ought to qualify for the Academic discount through your institution.
Just one thing about the clean-up (I am sure you know it already) – when you do any sort of image manipulation, please keep all the originals just in case the editor calls for them. The Microscopy Society of America has some guidelines which most journals adhere to.
Thanks, Kausik!
I’m a little afraid of jumping from PS6/7 to CS5 now, still. Also, does the Elements or Lightroom package do all that I mentioned above? There’s always infinitely more than I really need in there. Plus, some colleagues are surprised I do layout in PS and wonder why I don’t ALSO get Illustrator and do it there. Because I don’t know how, to start?
I would second your injunction to keep all originals – it’s an “of course” – which is why I like to also keep the reworked layer file as well, just to know what has been done.
I forgot to add, I like to render layers semi-transparent and do my own rotating and stitching if I have to, or superimpose my own fluorescent channels. Unless of course the new genius software does it all easily, in which case I could get used to it. We have a new fluorescent scope with software that does both on acquisition, as well as keeping the originals; very nice indeed.
Actually, having used all the versions since PS7, I can tell you that the learning curve for CS5 is pretty smooth. CS5 would, of course, add a host of capabilities that you may never need. The Elements/Lightroom package(s) is/are more suitable for photograph editing, but for scientific work, you need the regular (i.e. CS) version.
Illustrator is more useful if you work with vector graphics, including SVG. My wife uses a lot of Illustrator for her figures, while I use Photoshop. It may boil down to personal preferences. If you have bitmaps from the source, you cannot edit them in Illustrator, but can in Photoshop.
Photoshop CS3-5 handle layers beautifully. Usually, softwares associated with Fluorescent scopes are capable of exporting uncompressed, layered TIFFs. If that is the case with yours, you can absolutely rework them in Photoshop.
Okay – watch out, then, as I know on whom to call when I get stuck! Here comes my investment in legit software once more.
I have a young colleague who is using ImageJ to count her cells, but the new scope’s software handles that pretty easily now, too. In theory I wouldn’t mind supporting free and Open Source software, but in practice the learning curve can be very steep for someone for whom this is not the primary way we acquire data.
All the best. Don’t forget to get academic pricing, and if you can, go for the whole cs5 package instead of just photoshop cs5. Perhaps you cam download and use their fully functional demo for 30 days and see if it fits your needs? I think you’d love it.
Um, er, yeah… what Kausik said. 🙂
Elements – no, you’ll be frustrated with its lack of functionality.
Lightroom – no, designed more as a database for photographers, with good implementation for scripted presets, batch processing etc.
GIMP – powerful enough, but will drive you crazy.
Photoshop – yes, as Kausik says, will do absolutely everything you need.
I also agree with Kausik’s recommendation of getting the whole CS5 Creative Suite, which will also get you Illustrator. You may find it’s much easier to add captions and graphic elements than doing it in PS.
Good luck… I first started on CS2 and the initial learning curve was stupidly steep. I’m now on CS4 and the transition was dead simple.
One more thought – Illustrator also comes with a frustrating learning curve initially, especially if you’re used to moving stuff around in PowerPoint. But as with all Adobe products, once you get it going it kicks butt. <– that is the technical term
I’m worried about the stupidly steep learning curve, but the second opinion means a lot to me. Thanks very much! I’ll look into pricing – and the trial version should get me through the next couple of submissions with any luck in the meanwhile (sometime when I don’t have any more deadlines I’ll have to invent some).
Richard has rightly pointed out the learning curve, but I think that would be more for unknown functions you’d want to master afresh. Older functions (such as ones you are used to in PS6/7) should still be a breeze. Besides, there is now a thriving online community of amazing people using photoshop on a regular basis, and if you want help, you can always find it. If it is within my capacity, I’d certainly be glad to offer my assistance.
One comment on Richard’s message:
Especially for vector graphic elements. If you simply want to add a caption, you can use the Type Tool in PS and it creates a new layer for whatever you type in – a layer you can edit or modify at any time.
[Avowed PS fan here!!] 🙂
Are you tied to a PC?
With a Mac you could do most of the image manipulation in iPhoto (free with the Mac) and then assemble your figures in Keynote (Mac equivalent of PowerPoint – about 50 quid as part of iWork suite). Then export from Keynote as a pdf to keep the resolution and use Preview (free with the Mac) to convert to a bit-mapped file type (e.g. jpg. tiff).
Alternatively, a trick that gives you some more flexibility is to select all the elements on one slide, copy and then paste into a new file in Preview which you can save as a pdf. You can then re-save this file in a different format at pretty much any desired resolution.
This is how I make figures, though I don’t do microscopy.
what Stephen said. After switching to Macs from PCs I had the proverbial feeling that I’d been banging my head against a wall for years, and how nice it was to have stopped.
Thanks for your feedback!
I had been on Macs for years, quite a while ago. I am rather tied to PCs because of our image acquisition software. I can deal with either platform. Photoshop is pretty much the same on either. If only there was Photoshop for Linux (instead of Gimp) my life would be pretty much complete. Except for the image acquisition and other lab software (nucleic acid dosing, etc.). Sheesh.
“export from Keynote as a pdf to keep the resolution and use Preview (free with the Mac) to convert to a bit-mapped file type (e.g. jpg. tiff)”
I think that doesn’t really work. You lose a great deal of information along the way. I’d tried something equivalent recently when I was in a rush, but it is the transition of the original bitmap photo into Keynote or Powerpoint that is deadly. And jpg and tiff are not at all equivalent. Perhaps for your purposes that is adequate but for the kind of information I need to convey, almost exclusively microscopy, it isn’t.
Stephen, Henry – does it seriously keep all the resolution through the PDF conversion? PDFs are notorious for making graphics look, in a word, awful. Uncompressed formats (TIFF and the like) are much, much cleaner (and also give you the option of going to greater bit depths, which I’m 99% sure PDF will not do).
I like PDF for final wrap-up of documents I’m authoring that I don’t want changed, but I would never, ever use one to display a production-ready photographic image.
Richard,
One problem that I immediately see with Stephen’s suggestion is that “Export to PDF” functions are dependent upon the default settings of the export engine of the software concerned. I haven’t used Keynote (never felt the need to), but I do know that the default graphics export engine in Powerpoint is atrocious (One can raise it to a 300 dpi default by a registry hack, though). Therefore, when one exports to PDF from Powerpoint (and likely from Keynote, too), there is significant reduction/downscaling in resolution.
When I need to, I usually circumvent this problem by using Adobe Acrobat’s virtual printer function, i.e. I ‘print’ the documents to PDF and in the ensuing dialog box, I can put in whatever settings I want, ensuring that I can maintain the original resolution of my images. Acrobat can handle both TIFFs & JPEGs well. I am not sure if Preview has the same capabilities.
I’m not 100% sure, ricardipus and Kausik, but I think the detail of the image is maintained in the pdf. There are 3 settings — good, better, best.
Kausik – in my experience Powerpoint works much better for pdf or image export on a mac – resolutions up to 1600 dpi are allowed).
I don’t think there can be anything particularly special about microscopy images since I suppose they end up being published at about 300 dpi n the journal which is the resolution I work to in structural papers.
As far as specs go, that is correct for most journals, Prof. Curry. Generally, it is 300 dpi for color images in RGB or CMYK colorspace, 600 dpi for halftone images, and 1200 dpi for line-art graphics.
Once it’s a final image, at 300 dpi at the print size, then it doesn’t really matter which program generates the flat picture.
However, the issue is upstream. I need layers, I need to work at a higher resolution because I often crop my photos to best showcase what the reader is supposed to see in the figure. I need to rotate things and position others, and not be held to the rather coarse grid on Powerpoint. Above all, my impression is that when you import a high resolution photograph to Powerpoint, and then do transformations to it, those are destructive – you definitively alter the photo you have imported. So you lose information with each gesture, making it important to get everything right on the first go. Not my usual way of working.
The only thing about microscopy is that as much as we try to take photographs that are optimal from the outset, they usually aren’t, in particular depending on the publication to which we’ll be submitting the images. High resolution for a postage stamp-sized subfigure is not as necessary as high contrast, for example. A single cell in a large field won’t work at all, whereas in the old days I remember entire pages for single photomicrographs with multiple interesting bits of information in the field, in certain journals with the legend on the next page as a footnote.
Anyhow, I’ve got a reasonable quote and the trial version downloaded, and a bunch of new photos to play with.
1) Ooo, can you please tip me off as to this registry hack?
2) Thanks for the tip on printing to PDF – that might turn out to be most helpful. 🙂
EDIT – found it, thanks anyway:
http://support.microsoft.com/kb/827745
in case anyone else reading is interested. 🙂
Your Google-fu got the better of me… 😀 I have been meaning to send you that link for the past three hours, but couldn’t manage to sit down to do so. Sigh.
Anyway, try it and see if it works for you. Just FYI, this pertains to saving a whole slide as an image that you can later modify.
Just a note to say this registry hack works for me. And the CS5 version of Photoshop, too. I’m a happy camper.
w00t!