Well, when I wrote that long post about setting up food photography a couple of weeks ago, I promised a follow-up post that talked about post production on your images.
What do I mean by post production? No, it ain’t making fence posts. What I mean is everything that happens to the image after taking the shot. It is really pretty rare these days that I get a shot off the camera and am 100% happy with the shot, and just upload it straight to my blog. In fact, that honestly never happens. I always end up tweeking a bunch of things with the image until I get something that I reckon is OK to show.
OK details. What EXACTLY am I talking about? I am talking about cropping an image. I am talking about adjusting white balance, exposure, color adjustments, sharpness and all that jazz – all in the computer ater taking a shot.
I have been known to say “screw it, I can fix it in post”, and at work I have heard that from a good many people (I work as a computer graphics artist) too. The reality really is that most of the time only a certain amount can be done in post. So, you put complete crap in, you are going to get what we call “a polished turd” out. Sorry for the crudeness, just a term that a few of us fling around.
Most professional food photographers I know try and get it bloody close to perfect when shooting (before doing any post work). This is honestly the way to go. Try to get your white balance right in the camera. Try and get your exposure pretty much perfect before hitting post production. This will yield a better image. Trying to fix bad lighting after taking a shot is almost impossible. Only so much can be done in those nifty software packages.
Lets talk about software packages that you can use here for a second… There are a great number of packages out there now that are aimed at “developing” digital photographs. There are a few that I really like, but unfortunately they aren’t free. Thankfully most digitalSLR cameras come with free software that lets you edit your pictures. Canon (the camera brand that I shoot with) has Digital Photo Professional, which actually can do a pretty decent job of most adjustments. It is a little slow and clunky to use, but it gets the job done (some things really very well indeed), and it’s free. I am sure Nikon has something similar.
If you are looking at spending some money, then there are three that I recommend. First off is Adobe Lightroom. This is a great piece of software from the boys and girls that bring you Photoshop. Aimed at the professional photographer, it has a really slick workflow, is easy to use, and very responsive. I like it. Next up is Capture One Pro, from Phase One. This is another high-end piece of equipment, and really does a very similar job to Lightroom. If you look on most photography forums you will find heated debates about which of these two is better. I ain’t going there people, save the drama. Finally there is Photoshop. We all know photoshop. It has some half decent tools for adjusting photographs – but it does lack some of the controls of Lightroom or Capture One, uneless you really want to jump into some of the more advanced modules in Photoshop.
If you are thinking about really getting into food photography, I recommend one of those three. Either Lightroom or Capture One would be my first choice, however Photoshop is a handy thing to have around for doing other image tasks, like removing dust, stray bits of chopped parsley and so forth. Whatever you choose, the basics are the same – you have white balance/exposure/color adjustments in all of them.
I am not going to talk about workflow in a certain package. Heck, I am not going to even mention what software I use (it is one of those three..) – because I don’t want to bias anything.
To demonstrate my workflow, we are going to take one of my previous shots through the image post production pipeline – and natter about it on the way. The shot is from the salmon crudo post I did a little while back. I shoot in RAW format (talked about before in that previous post on food photography) – and the controls that I am talking about here are available for RAW images, and (depending on the package) JPEG too.
Here is what I am going to cover:
Getting a shot that is close to ideal from the camera
Histogram graph and exposure
GETTING A SHOT THAT IS CLOSE TO IDEAL:
I hate to admit this, and wished it wasn’t true sometimes.. Only so much can be fixed “in post”. We can only adjust exposure so much. Colors can be tweaked a lot, but you push it too far and you start degrading the image so much it starts to look odd. Been there, done that, got the t-shirt and mug, don’t want to do it again.
The idea is to try and make sure that the shot that you take on the camera is as correctly exposed as possible, and has the correct white balance. Thankfully most cameras these days have some tools on them that help you analyze images before you even download the image to your computer.
For white balance there are many options (for a more in depth talk about what white balance is, see my last blog entry on food photography). You can either set your white balance to the lighting conditions in which you are taking the shot (daylight, skylight, light bulb, halogen etc), or you can set a custom white balance. Most digitalSLRs let you set a custom white balance by taking a shot of something that is a neutral color (it doesn’t have to be white, but something that has no color – a very light gray is ideal). Personally, when doing this I put one of my white bounce boards down on my shooting surface, and use that. You then take a shot, and tell the camera to use that shot as “white”. After that, any shot you take should have pretty much the correct white balance.
I find that even when I do set the white balance correctly on camera, I often end up adjusting it slightly once I get it onto my computer.
Exposure. After you take a shot, you can preview it on the back of the camera (or on computer if you are “shooting tethered”). Most digitalSLRs and some compact cameras also give you tools to help see if an image is over or under exposed. These are really useful, and you should certainly familarize yourself with these tools. The most important of these is the image histogram, which I will talk about in a minute. Most cameras will also highlight areas that are over exposed (pure white) or under exposed (pure black). This can help you see if the actual food in the shot is bright enough, too bright, or too dark.
So, familarize yourself with the image information tools on your camera. Take a shot, and make sure the food is correctly exposed. Personally, I am not too woried if the background is a little over or under exposed, as long as the food is nicely lit and not too bright or dark.
Finally lets talk about the field of view – or just how much you have in shot when you take a picture. I tend to “shoot wide, crop later”. This means just what you think.. Take a shot a wider than you might want, since you can always crop out the bad stuff in your software package. This gives you more creative freedom, especially when leanring about composition. Some of my more favorite shots are actually crops of a much broader image. Cropping can certainly make an image much more dynamic and interesting – more on that later too.
On the left is the raw image out of the camera. On the right is the cropped shot. You can see that I shot the image a little wider than the final cropped image that made it onto my blog. My personal preference is for a reasonably tight crop to the food, with some of the food cropped out. The idea here is that if you don’t show absolutely everything, you leave a little up to the imagination. It is intriguing (hopefully), makes people think, and I reckon makes an image more interesting. Whilst I like the napkin under the plate, The original shot for me is too wide (especially for a small image for a blog) – at blog size it would be hard to see the different textures between the fish and vegetables, however cropping in a bit reveals a lot more detail, and I think makes for a more interesting shot.
The final crop at export resolution:
So, shoot a little wider than you might want, and you have some room to play with the crop composition later. That is never a bad thing in my book.
I talked quite a bit about white balance in my previous food photography post – so go there to check out the details. We can however play with the white balance in most photography image editing packages. Most packages let you pick a custom white balance by selecting an area of the image that should be a neutral color – that is one that isn’t red, green, blue etc, but one that is a pure gray (whether it is a light, dark, or white gray it doesn’t matter). I play with this a bit – I might pick a white that is on a plate, and see how that looks on the image. Sometimes it might make the image too blue or yellow. No problem, just pick a different area to use as a neutral color.
If that just doesn’t work too well, most software lets you manually adjust white balance with a slider (well, two sliders to be precise). My suggestion here is to use the color picker to get the white balance close, then use these sliders to fine tune the balance to your liking. The slider will generally change what is “white” from being very blue to very yellow. Just move the slider a bit towards the yellow if the neutrals look too blue, and the opposite if the image looks too yellow.
At first this dodgy little graph can be really confusing. The graph of your falling 401K value ain’t useful. The histrogram graph that shows your images exposure is the complete opposite – a joy to look at and extremely useful.
So, lets take a look at the histogram graph for this shot.
wow, that looks interesting doesn’t it.
But don’t shut down this browser window just yet, and shout that Matt is a complete bore. Lets look at what this image shows.
So, it is a graph that shows how bright or dark the pixels in your image are. Across the bottom of the graph it goes from dark to light. Up the graph is the amount of pixels that are that brightness (if that makes sense).
So if the left of the graph is dark, and the graph is really high over there, it must mean that a lot of pixels in that image are dark. If the right of the graph is white, and the graph is really high over there, it must mean that a lot of pixels are either light or dark.
The far left of the graph is pure black. The far right of the image is pure white. We can call pure black badly under exposed and pure white badly over exposed.
Personally, I like an image that doesn’t have too much pure black or pure white in it. So the goal of most of my exposure adjustments is to make sure that most pixels fall somewhere between pure white and pure black. We can thankfully use the histogram to see just how well we are doing at that.
Most software packages also have an option in the historgram to highlight both pure white and pure black in the image – this gives you a great visual feedback for what areas are over or under exposed. When you are fine tuning the exposure of your image, I really recommend turning on that feature if you can. I am not too bothered in areas around the food are over exposed, as long as the actual food isn’t too bright or dark.
Lets look at the graph again for a second. You can see that all of the pixels are sitting on the left 3/4 of the graph. This means that most of the image is somewhat dark – there is absolutely no part of the image that is bright – and certainly nothing that is pure white.
Here is the image again:
We can see that it is indeed looking a bit dark. That should be a white plate that it is sitting on, but it looks, well, dark gray. This is exactly what the histogram is showing – most of the image is dark – mid gray or below. Most image editing software has controls for adjusting exposure. We want to jump into those exposure controls now, and increase the exposure until we get a better exposed image.
On the left is the original shot, with its histogram below. On the right is the shot that has had its exposure corrected. You can see that graph now covers the complete range from black to white – and if you look really closely, it stops just before pure white, and just before pure black.
The plate now looks white. The food looks much more vibrant, crisp and alive. By looking at the histogram we can also be certain that we haven’t over-exposed the image – nothing is pure white.
My personal preference is for pushing the exposure a little further than most – I like a white, clean crisp image, but that is just my taste. Play around with image exposure and keep an eye on that rather boring, but rather useful graph.
When editing RAW images you have the option of adjusting either Exposure, or Brightness/Contrast. The exposure option is somewhat similar to how exposure compensation works on your camera, and should give a more natural adjustment than the standard brightness/contrast adjustments. Some photo editing software lets you make exposure adjustments to JPEG images too – but it won’t yield the same quality of adjustment as you can get with a RAW image, since some exposure information is actually automagically saved into the RAW file.
Now comes the highly controversial topic of color balance and adjustment. If everything is perfect on your shoot, then there should be no need for adjusting colors in post production. With me, nothing is quite perfect, especially since I don’t have much time to shoot the shots taken for my blog.
I have found there are some foods where the color just never comes through the camera quite perfectly. Red beets are one of those for me. They always end up looking too purple. Thankfully Photoshop, Lightroom and Capture One all have controls for adjusting individual colors. In Photoshop it is called “hue/saturation”. It is called something similar in Lightroom and Capture One. What these tools let you do is isolate a certain color range (just the purple or reds for instance with the beets), and tweek their color. We can make the reds and purples a bit more yellow – to remove that purple tone. We could also make them lighter or darker if we wanted to.
Salmon is somewhat of an orange/pink color. Want it to pop and jump off the plate? Just boost the saturation a little of the oranges and reds.
That is it really. When I first open up an image, the first thing I do is adjust the white balance. Until that is set correctly, it is really hard to adjust anything else. Next down the line is any exposure adjustment that might be needed, always using the histogram graph as an aid. Finally I go in and adjust color balance and saturation. The whole process can take anywhere from 2 minutes to 2 hours, depending on the image, and depending on how much time and love I spent on setting up and shooting the photo in the first place…. Time spent taking the shot is less time spent in post production.