Hi Folks,
I’m finally getting around to fixing the missing images in a serious way. I think most of the tutorial images are now present. I know there are still some missing tool icons and also some of the script download links may not work either. Please be patient, I will now start making a few passes over the site to fix up the broken links. Then I will ask for some help to update the tutorials to newer versions of the gimp.
Happy Gimping!
Gentle Readers,
I’ve managed to get most of the tutorial text migrated over, and preserving all the old links (I think). Unfortunately, the images are presenting a bit of a problem. The old site had all the images located under each tutorial. WordPress does not like to do things that way, it seems. It wants to put them all in folders by month. Since a lot of the images have the same name, there will be lots of collisions. I’m still trying to figure out if there is any way to guide wordpress to put the files in separate directories. If anyone has any good ideas please drop me a line, or comment on this post.
In any case, I ask your continued good patience while I work this out, and if you need to access any of the old material with images you can visit the Internet Wayback Machine. Try this URL (you’ll have to paste it into your browser bar, I think:
Welcome Gentle Readers!
I have converted GIMPGuru.org from a static html site into a WordPress hosted one.
I am slowly migrating all the content over. Of most interest, perhaps, are the set of tutorials that have been replicated on the Tutorials page. I’ve tried to preserve the links so that everything works as before. Please let me know if it doesn’t by commenting to this post.
My main reason for this change is that the new format and dynamic nature of WordPress will allow me to do some things with the site that I had wanted to do for a while now, but hadn’t got around to doing because of my busy day job and family life. You can look for these changes coming down the road:
Keep your eyes peeled for the changes. And thanks for stopping by!
Happy GIMPing,
–Eric
UPDATE: I know a lot of the image links are broken–please be patient!
Eric, Returning from a trip to a canyon in southern Utah my friend informed me that all of his digital pictures were marred by a fingerprint smudge on the lens that created a dull haze to a good 75% of all images right in the center. I took copies of the images to see if I could use GIMP to improve them. Out of the recesses of my spotty memory I recalled reading something on the web about removing haze. I connected the dots and realized atmospheric haze was very similar to the fingerprint haze in these photos. I only read a few image editing sites, your key among them, so I found your article right away. With a bit of trial and error I found some good settings for unsharp mask that dramatically improved the quality of the images. Seriously amazing.
On a few images I even applied the unsharp mask to the entire image then selected the oblong fingerprint smudge area, highly feathered the edges (150 pixels) then repeated the unsharp mask on this area (sometimes increasing the amount) to avoid over sharpening the unsmudged corners of the image.
While the images are still soft, they have been salvaged from almost useless to descent quality by use of this one technique (OK, I did the usual color level adjustment too).
A picture is worth a thousand words so I have attached before and after images.
Thank you for taking the time to share your knowledge of image editing, especially your GIMP tutorials.
– Kris
| Before |
|
||||||
| After |
|
Dear Eric: I desperately needed to make the background of my head shot out of focus, and I’d never worked with Photoshop or the Gimp before. The most experience I’d ever had messing with images has been cropping and adjusting the gamma and the contrast (so essentially, nothing).
So I downloaded and installed the Gimp, and even though some of the screens looked a little different, I managed to figure it out (I am using the Windows version). Thanks to your amazing tutorial on blurring the background, I DID IT! I am so proud of myself! So I wanted you to see for yourself that your efforts in making your web page are very useful and that you are helping people like me.
Thank you so much again!
Yours,
–Dominique
 |
||
| First I just want to say again how glad I am to find this group. I have to say everyone here is extremely helpful and it’s very much appreciated. Being a user of Photoshop in the past, I really got used to a function for cropping…and I would like to know how to duplicate it in GIMP. Basically, when I would click on the crop tool in PS, a toolbar would show up and give you a list of various sized to crop the photo to, and when you dragged the crop tool over the image, you could make it crop where you wanted AND be the size you wanted. You could also set the dpi in this same toolbar.
I am sure I am just not understanding how to do this in GIMP, and would appreciate it if anyone would be so kind as to explain it. I am not a technical type at all, so all that stuff about x and y confuses me when I try to crop with GIMP. I have other questions, but I will save those for another time. Thank you again. Lisa..who currently uses GIMP to edit digital photos of my 2 year old daughter =) |
| Hi Lisa,What you are talking about is setting a particular aspect ratio when you crop. This is done fairly straightforwardly in the GIMP, albeit perhaps not as simply as with the toolbar feature you describe. Here’s one way to do it…
Original photo. I’d like to make a 4×6 inch print (fairly standard print size at any US photofinisher) of a crop from it. |
 |
Choose the rectangular selection tool (keyboard: R) or ( ).If you want to crop to a certain print size (e.g. 4×6 in) or a certain aspect ratio (e.g. 2:3; same as 4×6) then in the Tool Options dialog, choose “Fixed aspect ratio” from the drop down dialog. Don’t choose “Fixed size” even if that is what you think you want.
In the Width and Height fields, enter the numbers for your desired aspect ratio. A 35mm film frame is 24mm x 36mm, so it has an aspect ratio of 2:3. This corresponds to common print sizes of 3.5 x 5 in (approx) and 4 x 6. An 8 x 10 print is related to the 4:5 aspect ratio of large format film, while 7×5 is closer to the 4:3 aspect ratio used by most digicams today. You can set the units field appropriate for your numbers (e.g. in this case I set it to “in”, since I want a 4×6 inch print), although I don’t think it’s that important at this stage. You do need to enter whole numbers, so round up to the nearest whole number aspect ratio. |
 |
Click and drag in the image window to marquee your selection. Zoom in a bit first if you like (as I did here). Don’t worry about getting the framing too precise, it is just a starting point. You should notice that the drag selection is constrained to your fixed aspect ratio. If you make a mistake just clear the selection (Shift+Ctrl+A) and drag again to make a different selection.Some handy keyboard shortcuts at this stage are:
Once the selection is approximately where you want it, choose the Crop tool (keyboard: Shift+C) or ( |
 |
In the Crop & Resize Information dialog, click the “From Selection” button. The crop lines will be set to your selection made in the previous step. Now press Shift+Ctrl+A to clear the selection.Now you can fine tune the crop. Some handy keyboard shortcuts are:
You can also use the mouse to enlarge or reduce the crop from any corner at this point, but be sure to hold down the SHIFT key or it won’t keep the aspect ratio. Once you’ve got the crop lines positioned exactly as you want them, click the “Crop” button in the Crop dialog. Don’t click Resize as this will only crop the pixels outside the crop lines but leave the canvas at the original size. If you make a mistake, just Undo (Ctrl+Z) and try again. |
 
|
| Finally, set the print size (Image/Print Size) and set the width and height to the desired size. The dimensions should be compatible with your aspect ratio chosen in step 1 and you also probably want to make sure that the resolution (in PPI) is sufficient for a good, sharp print. In this case I chose to go with 3.5 x 5.25 which is the same 2:3 aspect ratio as 4 x 6. At 3.5 x 5.25 I’m getting close to 240 PPI which is quite sufficient for a very sharp print. |  |
| Hi Everyone…I don’t know how I missed this group when searching for a GIMP support group. I am new to the program but I really like it. The problem is this…I am making a collage and I am trying to resize the pictures and paste them into the collage. I know the steps to make the collage in general, but the sizing of the images is driving me nuts.
I made a new 8×10 canvas. I resized 3 images (or so I thought I did) to 3×5. I have one image at 8×10 (again so I thought). I am trying to put the 8×10 image on the 8×10 canvas and lower the opacity to 60% . I then want to take each of the other 3 images and put them on the 8×10 image which is on the 8×10 canvas. When I check the image properties they are all the sizes I made them, but when I copy and paste them on the 8×10 canvas, they are still huge or only part of the image shows up. I tried to do a web search to find the answer, but I don’t even know how to title my problem. If anyone here has even a remote idea of how to help, please post. I’d really appreciate it. Thank you!! Lisa |
Hi Lisa,
Here is how I would go about it.
–Eric
| Open your “background” image. |  |
| Set the image print size (Image/Print Size).Choose a printing resolution that gives you approximately the size you want. In this case I used 240 PPI because I know that will give me a nice sharp print.
If you highlight one of the resolution fields, type a value and then hit TAB, you should see the width and height change. Remember the PPI you chose. |
 |
| If you are going to dial down the opacity, you really need something behind the image that will show through. If you want a kind of gauzy effect you can just put an opaque white background there.To do that I duplicated the Background layer, and renamed it “Main Image”. Then I selected the Background layer in the Layers dialog, chose the bucket fill tool from the toolbox, switched to a white foreground, selected all in the image (Select/All) and clicked in the image to fill the Background with white.
I then selected the Main Image layer and dialed the Opacity of the layer down to around 75%. Finally, add a new layer on top of the image using the new layer button in the Layers dialog (this will be for the overlaid images). |
  |
| Open up another image that you want to place on the main image to form the collage. Complete any image editing work you want to do on it first.Choose Image/Scale Image (one of the nice things about the GIMP is that it uses different dialogs for resampling and setting print resolution, something that confuses so many folks in Photoshop because it only uses one dialog for both).
Set the units that you want the resulting size of the small image to be (in this case I used “inches”). Put in the same resolution that you used earlier (in this case 240). Set interpolation type to “Cubic” for best quality resampling. Now dial in the size of the longest dimension you want the small image to be. In this case I wanted 3 inches wide, so I highlighted the width parameter, typed in 3 and hit TAB. The other dimension will be calculated. Underneath the width and height you can see the dimension that the image will be in pixels after downsampling (here 720×540). Click “Scale”. The image will be downsampled. A sharpening step here would be wise (Filters/Enhance/Unsharp Mask). In this case I didn’t bother. Finally, Select/All and Edit/Copy. |
  |
| Go to the collage image and Edit/Paste. Select the Move tool from the GIMP toolbox and move the image to just exactly where you want it.In the Layers dialog, anchor the image.
Voila! Repeat the last step with other images to be pasted. This is a pretty quick and unpolished example, but it should convey the general idea. You could obviously do lots of fancy things to the small images (borders, drop shadows, etc.) before downsampling and pasting them into the collage. |
  |
| Good morning, Guru: Thank you for the excellent tutorials.
Have you written one on editing B&W photographs? I have several photos of family members. Several of the photos are over 100 years old. I have already scanned them into Gimp. They are replete with scratches and are very fuzzy. Any pointers you may willing to give me would be greatly appreciated. I am not a professional photographer, I am a Senior Computer Systems Engineer, and am still somewhat trainable. Thanks, in advance. Ron |
Wilbur |
| Hi Ron,Sure, there is a lot you can do to clean up old photographs. Before we dive into it though, I have to mention the obvious, in that you are somewhat limited in what you can reasonably achieve by the quality of the input (original) image.
In this particular example, the “fuzziness” you refer to can be improved, but not to the extent of a clearly focused, sharp photograph. The blurriness of this photo could be due to any number of factors including subject movement during a slow shutter speed (the image being taken with available window light, it would seem), poor focus, poor lenses, poor scanner focus, etc. That being said, there is a lot we can improve on, so let’s get started! [Aside: you can often get a better starting point from your scanner. Check if there is an option in the scanning software for “old B&W print”. Also, investigate the infrared scratch and dust speck removal features of your scanner (e.g. “ICE” on Nikon scanners); this can give you a much better starting point.] The first order of business is to crop out that ugly border using the crop tool ( |
 |
||
| The next step is to convert this to B&W from the current old, faded sepia. You can follow the instructions in this tutorial to figure out the most pleasing conversion. After examining the separate RGB channels, I settled for a standard mode change to Grayscale (Image/Mode/Grayscale). Then I switch back to RGB mode (Image/Mode/RGB) so I have more filters to work with as well as options for re-toning. |  |
||
Now we are going to remove scratches and dust with the clone tool. There is some useful instruction in this tutorial. Basically, you want to use the Clone tool ( ), with a small soft brush; try turning down the opacity in the Tool Options dialog box. I’ve circled the areas that need the most work.Dial down the opacity on fine details, such as hair. The cleaned version is shown at right. We could spend a long time cleaning this image, but since it is old some dust gives it character. If you have an infra-red dust filter on your scanner (e.g. “ICE” on Nikon scanners) this can prevent a lot of cloning work. |
  |
||
| Next we’ll attack the dynamic range, or lack thereof. Scanners will typically leave a “haze” on scanned images that can be rectified by a simple levels adjustment to normalize the image and increase the apparent contrast. So open the levels dialog (Layer/Color/Levels). Notice how the histogram is compressed, and not spanning the full tonality that is available. Bring the black point up and the white point down (hitting “Auto” will usually also do a reasonable job, although manual adjustment gives more control). |  
|
||
| At this stage let’s retone the image, using this technique. For this photo I prefer not to go back to sepia, but rather something like Platinum (01). |  |
||
| Finally, we’ll sharpen the photo as best we can using smart sharpening redux, in order to avoid oversharpening the scratches, dust spots, etc. Here is the edgemask I made, the sharpening parameters used, and the result.I used lower parameters than usual, because the image (as sent to me) was small, and therefore requires less sharpening, but also because the image is so blurry to begin with. We can never hope to overcome this level of blurriness by digital sharpening. What happens instead is that some (bold) edges (e.g. the side of the light face against the dark background, the seams of the overalls on the chest) appear very sharp while finer transitions (in the eyes, hair) are still hopelessly blurry. It creates an incongruity for the eye, and actually accentuates sharpening artefacts that would normally not be so objectionable. |  
|
||
| And here’s our end result, side-by-side with the original. To my eye, it’s an improvement, although I would consider skipping the sharpening step for the reasons I outlined above. |
|
| Dear Guru, I got a new Nikon D70 3 days ago. My wife is quite upset about my “expensive toys” and tries to fault digital cameras in general and the D70 in particular whenever possible.
She says: but can it do B&W. I say: Sure. I take a photo of my favorite driftwood (wood.jpg) and convert it, as suggested by your tutorial (ch. mixer), and get (wood-bw-o.jpg).
This obviously is greeted with lots of scorn. Look at (pier.jpg) (A scan of a print I made from a B&W film negative), she says. This is closer to how a B&W should look (Of course she thinks a screen is no good either, only a print is, but that’s another story.) I tried Pushing the contrast way up, brightness a little down, then added some noise (Filter/Noise/Noisify) to get some grain, and get something like (wood-bw.jpg)
My wife (An artist, painter, BTW) is not happy with that too. Any enlightenment will be greatly appreciated and might stop a divorce Thanks, Joseph |
|||||
|
I love this question! Your wife sounds like a lot of folks in our local photo club here–“pooh pooh digital” no matter how good the image looks. I’m not sure that you can do about that, vis-a-vis the divorce. How about a nice candlelight dinner for two, instead? I like your picture. The highlights are just slightly blown out, and after conversion to B&W it just accentuates that a bit, but it is not overwhelming by any means, just wanted to draw it to your attention. When faced with a B&W conversion, I always start by decomposing to separate R-G-B images to see what each channel has to offer (this is a good step before channel mixing anyway, so you know what you are mixing). After doing this with your color image, I liked the blue channel the best. Often the blue channel is noisy, but with your new D70 it was very clean. The blue channel often has the best contrast, so if you are looking for a contrasty B&W like the pier image, blue is a good base. After playing with channel mixing a bit, I felt that the green and red channels didn’t really add anything interesting, so I discarded them. I then ran an unsharp mask on it with something like radius=1.4, amount=.26, threshold=4 to highlight the lines in the image, since that is a big part of what makes it work. The result is (try1.jpg). I would stop here, speaking for myself. The result is a little more contrasty than your first B&W attempt, but not so much that the shadows are all blocked up like your second attempt.
I tried adding grain, as described in my tutorial here, but I decided that it didn’t really add to this image, and really made it worse. Grain can help add interest to large smooth areas of tone, but your image has lots of texture already. The silky smooth digital capture works to your advantage here, in my opinion. If you really want to boost the contrast further, try a “local contrast enhancement”. The basic idea is to run unsharp mask with an absurdly high radius, low amount, and zero threshold. The result is shown in (try2.jpg) It’s a little too over the top for me, but see if your wife likes this any better. Best Regards,
|
) and click on the image.
GIMP Guru 