This is another of the grafittishots I took last week (I have two, maybe three left that I’ll put up at some point) and is the only one of all of them that displays an entire piece. What I like about all of these shots is the vibrancy of the colour, which is mostly a consequence of the artists using metallic paint for portions of their work. But what I like about this one is the grass and weeds at the bottom of the image. For the rest of the shots I focussed much more sharply on the work itself, this one contains a bit of the surrounding scene. And I guess that the reason I like it is to do with the contrast between the sharp industrial feel of the graffiti and the contrasting softness and delicacy of the foliage.
Alternatively, it may just be that the greenery provides a bit of balance ot the orange/red/yellow of this piece of work ;-)
camera capture date aperture shutter speed shooting mode exposure bias metering mode ISO focal length image quality white balance
Canon G5
2.14pm on 30/4/04
f2.8
1/250
aperture priority
+0.0
evaluative
50
8.0mm
RAW
auto
It is for this vibrancy in colour that I spent long hours around my photos in Photoshop. Can you tell us more about how did you achieve it here?
Is that green on the leaves very far from the original?
comment bybrad bane at 01:59 AM (GMT) on 7 May, 2004
love the color. its always nice to see art like this to brighten up a drab city.
comment by djn1 at 11:02 AM (GMT) on 7 May, 2004
PCG: I'm a bit short of time at the moment, but I'll put up some example images some time over the next few days that show how this image looked at the various stages (RAW image, Levels adjustment, Curves adjustment, Digital Velvia plugin). As for your question, no, it's no a million miles away in terms of colour - just less saturated and quite flat in comparison. I'll drop you an email when I've had a chance to put the various images up.
The colors are very nice. Graffiti can lighten up dark and boring walls in cities. But also ruin a lot of fine buildings...Always something negative about good things. Your camera is very good at catching up the colors!
comment by djn1 at 12:52 PM (GMT) on 7 May, 2004
PCG (and anyone else who's interested): I've produced a sequence of images (with descriptions) that go through the various stages of how this image was produced at the following url:
I checked out your PS workflow examples and I'm just curious about how you use the Unsharp mask. I've seen lots of ways to use it and I'm wondering if you use a certain set of values for the Amount, Radius and Threshold or if it varies for each photo.
comment by djn1 at 08:33 PM (GMT) on 7 May, 2004
twb: I'm thinking of putting a more extensive workflow example up at some point, including something on Usharp Mask, but here's a brief summary:
1) Reduce the image to the required size; i.e. NEVER sharpen an original. All forms of sharpening are really nothing of the sort; i.e. they can't add in detail that isn't there in the first place. Instead, they give the appearance of sharpness by exaggerating the edge contrast. You will have noticed this if you’ve sharpened an image that contains (for example) a dark edge against a medium density background (e.g. power cables against a fairly dark sky) that at a certain level of sharpening (and above) you introduce a halo at the edge of the dark object. At this point the sharpening looks artificial.
2) Anyway, as for the workflow: convert your image to LAB colour and select the lightness channel (which specifies the brightness of a given pixel independently of the colour information). The reason for doing this is that if you work on a full colour image the sharpening process can change the colour values as well as the apparent sharpness, which can end up looking artificial.
3) Following this I tend to use the Unsharp Mask twice in succession with the following settings: on the first pass I use an amount of between 120 and 250%, a radius of 0.3 pixels, and a threshold of 0. On the second pass I up the amount to something between 250 and 400%, a radius of 0.2 pixels, and again, a threshold of 0. The reason for using it twice is that it seems to produce a more subtle result than trying to do it all in one go.
4) Convert the image back to RGB.
5) At this point you may be finished but I often find that there’s the odd haloed edge here and that some hard diagonal edges may appear a little jagged, both of which don’t look particularly pleasant or natural. So, using a soft brush of around 4 to 10 pixels, with an opacity between 25 and 40% I tend to blur both of these until they still look sharp, but not artificially so.
If anyone does anything radically different to this I'd be interested to hear. Also, if you can suggest any improvements, likewise, I'd be grateful.
PS – ok, so this wasn’t all that brief after all ;-)
Dave: I used your Unsharp Mask workflow on the photo I just posted to my website (and I swear it's merely a coincidence that it's also graffiti subject matter!). This workflow is vastly different from others I've learned and I LIKE it! Thanks so much for sharing and enjoy your weekend.
comment by djn1 at 12:08 AM (GMT) on 8 May, 2004
twb: thanks, and thanks for the link on your site - it's a great shot by the way :-) You mentioned that the sharpening workflows you've learned are vastly different from others you've learned - if you've got the time could you explain them? The one I use is an adapted from one I came across in one of the many digital photo magazines (I can't remember which one now) and is basically a scaled down version of their suggestions (i.e. 0.3px rather that 1.8px for the first pass, etc). I figured, given that they were talking about full resolution images for printing, that the radius would need to be much smaller for onscreen work. And it seems to work ok, but I'd still be interested in any other suggestions.
This is another of the grafitti shots I took last week (I have two, maybe three left that I’ll put up at some point) and is the only one of all of them that displays an entire piece. What I like about all of these shots is the vibrancy of the colour, which is mostly a consequence of the artists using metallic paint for portions of their work. But what I like about this one is the grass and weeds at the bottom of the image. For the rest of the shots I focussed much more sharply on the work itself, this one contains a bit of the surrounding scene. And I guess that the reason I like it is to do with the contrast between the sharp industrial feel of the graffiti and the contrasting softness and delicacy of the foliage.
Alternatively, it may just be that the greenery provides a bit of balance ot the orange/red/yellow of this piece of work ;-)
capture date
aperture
shutter speed
shooting mode
exposure bias
metering mode
ISO
focal length
image quality
white balance
2.14pm on 30/4/04
f2.8
1/250
aperture priority
+0.0
evaluative
50
8.0mm
RAW
auto
It is for this vibrancy in colour that I spent long hours around my photos in Photoshop. Can you tell us more about how did you achieve it here?
Is that green on the leaves very far from the original?
love the color. its always nice to see art like this to brighten up a drab city.
PCG: I'm a bit short of time at the moment, but I'll put up some example images some time over the next few days that show how this image looked at the various stages (RAW image, Levels adjustment, Curves adjustment, Digital Velvia plugin). As for your question, no, it's no a million miles away in terms of colour - just less saturated and quite flat in comparison. I'll drop you an email when I've had a chance to put the various images up.
The colors are very nice. Graffiti can lighten up dark and boring walls in cities. But also ruin a lot of fine buildings...Always something negative about good things. Your camera is very good at catching up the colors!
PCG (and anyone else who's interested): I've produced a sequence of images (with descriptions) that go through the various stages of how this image was produced at the following url:
.../iblog/archives/ruby_deam.php
Thanks for the 'post production' examples you made--I also recently took some graffiti shots that could benefit from similar enhancements.
Wow, this photo just pops out at you!
I checked out your PS workflow examples and I'm just curious about how you use the Unsharp mask. I've seen lots of ways to use it and I'm wondering if you use a certain set of values for the Amount, Radius and Threshold or if it varies for each photo.
twb: I'm thinking of putting a more extensive workflow example up at some point, including something on Usharp Mask, but here's a brief summary:
1) Reduce the image to the required size; i.e. NEVER sharpen an original. All forms of sharpening are really nothing of the sort; i.e. they can't add in detail that isn't there in the first place. Instead, they give the appearance of sharpness by exaggerating the edge contrast. You will have noticed this if you’ve sharpened an image that contains (for example) a dark edge against a medium density background (e.g. power cables against a fairly dark sky) that at a certain level of sharpening (and above) you introduce a halo at the edge of the dark object. At this point the sharpening looks artificial.
2) Anyway, as for the workflow: convert your image to LAB colour and select the lightness channel (which specifies the brightness of a given pixel independently of the colour information). The reason for doing this is that if you work on a full colour image the sharpening process can change the colour values as well as the apparent sharpness, which can end up looking artificial.
3) Following this I tend to use the Unsharp Mask twice in succession with the following settings: on the first pass I use an amount of between 120 and 250%, a radius of 0.3 pixels, and a threshold of 0. On the second pass I up the amount to something between 250 and 400%, a radius of 0.2 pixels, and again, a threshold of 0. The reason for using it twice is that it seems to produce a more subtle result than trying to do it all in one go.
4) Convert the image back to RGB.
5) At this point you may be finished but I often find that there’s the odd haloed edge here and that some hard diagonal edges may appear a little jagged, both of which don’t look particularly pleasant or natural. So, using a soft brush of around 4 to 10 pixels, with an opacity between 25 and 40% I tend to blur both of these until they still look sharp, but not artificially so.
If anyone does anything radically different to this I'd be interested to hear. Also, if you can suggest any improvements, likewise, I'd be grateful.
PS – ok, so this wasn’t all that brief after all ;-)
thanks for the tips -- i never fail to learn something new when i come here. and your pictures of course are amazing. :)
Dave: I used your Unsharp Mask workflow on the photo I just posted to my website (and I swear it's merely a coincidence that it's also graffiti subject matter!). This workflow is vastly different from others I've learned and I LIKE it! Thanks so much for sharing and enjoy your weekend.
twb: thanks, and thanks for the link on your site - it's a great shot by the way :-) You mentioned that the sharpening workflows you've learned are vastly different from others you've learned - if you've got the time could you explain them? The one I use is an adapted from one I came across in one of the many digital photo magazines (I can't remember which one now) and is basically a scaled down version of their suggestions (i.e. 0.3px rather that 1.8px for the first pass, etc). I figured, given that they were talking about full resolution images for printing, that the radius would need to be much smaller for onscreen work. And it seems to work ok, but I'd still be interested in any other suggestions.