Category Archives: Techniques

Balloons and Background Lights

Most of the time catalog photography is pretty straight forward. White background, light the product, click, you’re done. Occasionally though I get a product that throws me for a loop and requires some trial and error, and a lot of frustration before the answer reveals itself.

Such was the case of this necklace I was photographing last week. This was part of a batch of about 70 pieces of jewelry, and I’ve shot over 500 pieces for this company. Usually, for necklaces, I hang them in front of a translucent panel of plastic that is backlight to give me a pure white background. By hanging the necklace a few inches in front of the panel, I am able to not blow out the edges of the necklace, which is critical when they are silver.

So my first attempt at photographing this necklace looked like this.

The client came back and said they needed the necklace to be photographed showing the whole necklace, and draped as it would be as it was worn. Ok, I figured I could lay it flat and photograph it that way. However, this opened up a can of worms that had me pulling my hair out for a couple of days.

The first problem was that the necklace wasn’t made to lay flat and it was similar to the problem of trying to make a flat map of the world. Complicating that issue was that I was working with pre-production pieces and the beads were tied so tight that they tended to kink, and the client wanted them to look smooth. Laying flat on a light table, I tried for an hour to get everything to lay correctly. But try as I might, as soon as I would try to smooth the second string, the first string would get jostled and I would have to start over. Short of super gluing the beads in place I didn’t think I would ever get it to work.

The second problem was that by having the necklace right on the light table, by the time I got enough light to make the background pure white, the ruby beads were glowing and the silver beads were blown out.

The third problem was that there were four similar necklaces, and the client wanted them to be photographed as identically as possible.

The final problem was that the client needed the necklaces back the next day to ship them back east which only added to the pressure.

I went home that night trying to think of a solution. I have a mannequin in the studio and putting the necklace on the mannequin I could see how it looked and it did hang much better, but there was no way to photograph the entire necklace that way. What I needed was something that would allow the necklace to drape that way but not have a neck. I thought about building some kind of form out of styrofoam or plaster, but as I needed to return the necklace the next day that left little time to build any kind of prop.

Then I thought about using a balloon, and if I could find a large white balloon, it would work for diffusing the light as well as supporting the necklace. So the next morning I went to Walgreens and hit pay dirt with a white punching balloon. Returning to the studio, I set the background light on short stand, and pointed it directly up. I attached a snoot to the front, not only to provide some support for the balloon, but even with the modeling light off, I didn’t want the head of the flash to explode the balloon. The balloon was taped to the snoot, and then the first necklace was laid on top of the balloon.

Not only did the balloon provide the perfect curved surface for the necklace, its non-slip surface made it a lot easier to gently nudge the strands where I wanted them without moving the neighboring strands. Once I had the first one in position the way I wanted it, I marked the top, bottom and sides with tape so I could lay the other three in a similar pattern.

The camera was mounted on a 10 foot ladder, and I shot directly down on the balloon.

Then it was simply a matter of taking two shots. The first shot I had the background light on, in addition to the softbox that was the key light. This blew out the background so I would be able to make a mask for the second shot.

The second shot, I turned the background light off and only used the key light, which gave the right color to the rubies and protected the edges of the silver beads.

Then using the mask created from the first shot, I applied it to the second shot. I still had to so some manual masking in Photoshop for the silver beads because the contrast level between the silver beads and the background made it hard to find the edges.

Here is the final shot sent to the client.

Restoring Antique Tintype Photos

Tintype photographs are the easiest type of antique photograph to identify. Despite what the name suggests, the substrate was usually a thin sheet of iron. The tintype process was patented in 1856 by Hamilton Smith and as the photo below shows, the tintype process was still being used at the turn of the century. Dating the clothing in the photo puts this image’s date around 1900 to 1905.

Scanned with scanner defaults

A tintype is a very underexposed negative that is on top of a black varnish finish. Because it’s a negative, the areas of the image that recorded shadows are clearer and show more of the black background. Conversely, the areas of the image that recorded highlights have more silver and show less of the black background. But since silver isn’t as bright as the white paper we’re used to seeing as a film substrate, tintype images tend to be dark and they don’t have a lot of contrast.

Therefore, when making digital copies and restorations of tintypes, the most important step of the process is getting a good scan of the original. This is one case where you do not want to use your scanner’s default settings. Doing so will get you an image that looks exactly like what your eyes see.

Scanned with adjusted settings

Instead, we want to look at the scanner’s preview image and set the histogram so that the black point is the darkest part of the image and the white point is the lightest. For most tintypes, this means you’re going to be using less than half of the contrast range your scanner provides. Because of that, it’s important to use the highest bit depth your scanner offers.

For example, if your scanner has 16 bit scanning depth, it can record 65,536 levels of gray. But due to the lack of any light grays and whites in a tintype, the values in a tintype will only use the lower range of that 65,536. So we may only be left with 20 or 30 thousand values to work with. It’s important to get as many values as you can in the scanning step so that when you get to the Photoshop work later, you have the most amount of information to work with.

When the image is scanned those 30,000 levels are going to be spread out over the full range of 65,536 levels. So all along the range there will be holes. If those holes are too big, you’ll notice definite steps in gradations of gray levels.

In the image shown here, there were numerous spots and scratches, way too many to fix by with the clone brush or the healing brush. Photoshop has a Dust and Scratches filter you works quickly, but it works by blurring the image. Since tintypes are usually pretty soft to begin with due to the technology of lenses back then and the small size of most images, it’s not desirable to make them even softer. What I did with this image was to duplicate the image on a new layer, use Dust and Scratches on the duplicate layer, and then set that layer to darken. That way the only pixels that were affected were the missing pixels.

Final Restored Portrait

That fixed about 80% of the problems with this image. After that I used the healing brush and the clone brush to fix a few other areas.

The final step was adding a Contrast adjustment layer with a slight S curve, just to deepen the shadows a bit.

Thinking Upside Down

The last few weeks I’ve been shooting the latest jewelry collections for Baroni Designs. These are all product shots to be used in their catalog and on their web site. Like most catalog shots, these are all shot on a white background. In the case of the products that hang, like earrings and necklaces, the background is a translucent piece of plastic that gets lit from behind with a soft box.

Most of the items are straight forward but every so often, as with any group of products, there’s one that throws you for a loop. In this case it was literally the loop that threw me.

For most of Baroni’s earrings I simply insert the ear hoop through a small hole in the plastic, just as if you were inserting it into your ear to wear it. The earring hangs as it naturally would. But for this one earring set, the loop itself is a large part of the earring’s design and it needed to be shown from the side. That ruled out the normal method I use.

I thought about laying it on a table, but the pearl, being much thicker than the wire would cause it to bend and not look like it was hanging.

I also thought about rigging it from some fishing line or a small hook and then Photoshopping it to remove the suspension device.

But by thinking upside down, I solved the problem in a way that looked natural and didn’t require any Photoshop work. I used a small piece of earthquake putty on the back side of the pearl and then stuck it to my piece of translucent plastic so it was hanging upside down. To get better separation of the white pearl from the white background, I surrounded the pearl with bits of black gaffer’s tape. Then I shot it as usual, with one soft box behind the plastic, and one soft box in front. I also used a small reflector card to control the specular reflections.

Converting color images to black and white in Photoshop

In a recent forum discussion, the question was asked what’s the best way to convert color images to black and white using Photoshop. As with almost any task in Photoshop, there are a number of ways to accomplish it and while some are personal preference, some methods do offer advantages to creating better images.

First, I’m going to define the “best” way as the method that allows you to fine tune the image the most. While some of the methods are easier and quicker than others, they provide little control over the final image.

There are 5 methods that I’m going to review and I’m going to present them in the order of least desirable to most desirable. They are:

Hue/Saturation Adjustment Layer
Image>Mode>Grayscale
Gradient Map Adjustment Layer
Channel Mixer Adjustment Layer
Black & White Adjustment Layer

There are two other methods that I’m not going to cover. First is making the conversion to the Raw file before it gets into Photoshop. Converting the raw file can produce better results in some cases, but the Black & White Adjustment in Adobe’s Camera Raw works similarly to the one in Photoshop. The second alternative is using a plugin which might give you even more options. But since I am happy with the amount of control that the Black and White Adjustment Layer provides, I haven’t explored purchasing a conversion plugin.

To help demonstrate the results of each of the 5 methods, I first created a sample file made up of 7 gradients. The first three gradients go from black to one of the primary transmissive colors (red, green, blue) and then to white. The next three gradients go from black to one of the primary reflective colors (yellow, cyan, magenta) and then to white. The last gradient goes from black to white using shades of gray.

Then I used the different conversion methods to see what effect they had.

Hue/Saturation Adjustment Layer

As you can see, reducing the saturation treats every color the same. You’re simply getting rid of the color. There are no options for this method. The black and white gradient is a bit different, and that might be due to the way I created it.

Image>Mode>Grayscale

I was actually surprised by the result of this method because I expected it to be identical to reducing the saturation to zero. Especially when you see the warning message that Photoshop throws up.

It sounds like it’s going to do the same thing that the saturation adjustment would do. But this method uses some algorithm that treats colors differently. This is just a setting that some Adobe engineer decided looked good for most images. It might look good for your image, then again it might not. But like the saturation method, you have no control over the result.

Gradient Map Adjustment Layer

Using a Gradient Map to convert to black and white is a method that I wasn’t familiar with until reading about it in the forum discussion mentioned earlier. So I looked it up on the web and found an article on how to do it but the author wasn’t sure exactly what it was doing. I don’t know either. It does provide a different result than the prior two methods. It also offers one level of control. You can change the softness value in the gradient to get slightly different results. Here are two conversions. The first one was converted with a softness value of 100% and the second was a softness value of 33% which the article recommended.

Softness set at 100%

Softness set to 33%

The effect of using a gradient map is pretty close to that of the grayscale method, except that it seems to darken the blue channel more. The effect of changing the softness level is subtle at best. So even though the method has some control, it’s not related to anything that you’re trying to accomplish in the image. Which gets me to the next method.

Channel Mixer Adjustment Layer

33% Red, 33% Green, 33% Blue

Photographers who have actually shot black and white film know that the best tools to improve the photo are contrast filters, a set of colored filters. By restricting the colors that hit the film, the photographer can change the contrast of the image. Since a black and white image can only convey information from contrast, this is very important. The channel mixer method allows you to use some of the control that contrast filters do for film. The advantage that digital has over film though is that the decision of which filter to use can be made later. You can even use one “filter” on the sky, and another “filter” on the foliage, something totally impossible in the film world.

For example if you want to lighten the leaves on the trees so they stand out better, you would increase the green channel. The example above shows all three channels being set to 33%. This gives a result identical to the saturation method.

But the wonderful thing is each of the three channels can be tuned from -100% to +100%. When I was using Photoshop CS and before, I used the channel mixing method and as a starting point for most images I used settings of Red 60%, Green 40%, and Blue 0%. This difference is illustrated in the image below.

I’m not saying those are the best values to use. But for the camera that I was using at the time, and the type of images that I take, that seemed to be a good starting point.

Black & White Adjustment Layer

When I upgraded to Photoshop CS3, I found there was a new conversion method available under Adjustment Layers called Black and White. Similar to the Channel Mixer, it allows you control the mix of colors, but instead of just the transmissive primaries available in Channel Mixer, it also has sliders for the three reflective primaries. I now use this method for my black and white conversions. Most of the time I make adjustments to the red and yellow channels. But once in awhile one of the other channels becomes really useful. For example if you’re trying to make a purple shirt show up better, sliding the magenta higher can help. The six controls allow you to target individual colors better.

This image shows the output of the sample file with the settings at each slider set at 50%. This gives you the same output as the reduce saturation method.

Here is the difference when you change the magenta to -200% and the cyan to +300%.

Finally, here is the output using the default values of Red 40, Yellow 60, Green 40, Cyan 60, Blue 20, Magenta 80.

Conclusion

While all of the methods will get you black and white image, if you want to control how the final image looks, the Black and White Adjustment Layer provides the most control, and I think it is also the most intuitive. Remember also, that in addition to the color conversion, your black and white image may need further changes to the contrast, either globally or locally. Global changes can be made with a Contrast Adjustment Layer. I usually darken the shadows and sometimes I lighten the highlights as well. Local changes can be done through dodging and burning. Like color conversion, there are numerous ways you can dodge and burn in Photoshop. That’s an entirely different subject to explore.

Here is a summary then of the different methods.

Hue/Saturation Adjustment Layer – Just throws the color info out. No user input. Every color adjusted equally.
Image>Mode>Grayscale – Color info removed, but Photoshop applies it’s own algorithm of what it thinks is best. No user input.
Gradient Map Adjustment Layer – It does something similar to Method #2, but not sure what it does. Minimal non intuitive control
Channel Mixer Adjustment Layer – 3 channels x 200 levels = 8,000,000 options and intuitive control interface
Black & White Adjustment Layer 6 channels x 500 levels = 15,625,000,0000,0000,000 options and intuitive control interface

If you’ve used method 5 and have also set the global and local contrast of your image, but you’re still not happy with your black and white image, then maybe it’s an image that just isn’t going to work as a black and white. Maybe it needed different lighting and/or composition to begin with to make it an effective subject in black and white.

Taking Professional Looking Self Portraits

With the rise of social networking, digital cameras and camera phones, there has also been a huge increase in the number of self portraits that people take. Unfortunately, photos from arms length are about the most unflattering way to take a self portrait because the wide angle lens needed for that close distance really distorts the facial features of the subject.

Lately I’ve been playing around in the studio doing some self portraits, mainly to have something fun to post on Facebook. I’ve done a hillbilly photo, several experiments using gels over the lights, and most recently, a pirate photo for Talk Like A Pirate day. Actually that one ended up not being a self portrait, because for the look I was trying to get, I was having trouble getting the right focus. So since a friend stopped by the studio I had him focus the camera and snap the pic.

But it got me to thinking there should be a way to carefully focus and frame the camera for self portraits. First let’s look at what doesn’t work.

1. Just use a small aperture and let the large depth of field (DOF) take care of it. Great unless you want to throw the background out of focus as I did in this shot.

2. Put the camera into auto focus mode. That works great if you’re there to put the focus point directly over the closest eye. But it’s very hard to do when you don’t know where that eye will end up in the frame. You might end up auto focusing on your nose, which is not a good thing when working with large apertures.

So after giving it some thought I looked around the studio to see what I could use as a placeholder for focusing purposes. My first attempt I used the head and torso of a mannequin. That didn’t work very well because she didn’t sit up properly.

The solution that I found that worked was to use a microphone boom stand. I set it so the knurled knob was directly in front of my eyes. Then I went to the camera and focused on that.

Next I set my cameras intervalometer so shoot 10 frames, just for good measure. With a ten second delay before the first exposure I had time to get back to the chair, get my pose adjusted just right and then swing the mike stand out of the way. For each of the pops, I moved my head forward just a fraction of an inch.

Then it was just a simple matter of opening the series in Bridge and checking to see which one had the absolute best focus on the eyes.

Another advantage to using this method is that it gave me a better target to frame in the camera. Usually I’d give myself lots of extra padding around the frame, figuring I could just crop into the photo since they were only going up on the web. But by doing it this way I could crop it correctly in the camera just as I would for a client.

The lighting set up I used can be seen in the first photo. It consists of the following:

Main Light: Photogenic 1250 DR with a 24″x32″ LiteDome soft box
Fill Light: 40″x80″ silver reflector
Background Light: Photogenic Studio Max III 160 with a 7″ reflector
Hair Light: Photogenic 300 DR with a 7″ reflector and barn doors
Kicker Light: Photogenic 1250 DR with a gridded 15″x50″ HalfDome softbox

Here is the finished self portrait:

Painting with Light

Over ten years ago I first came across the technique of painting with light when one of my web design clients, Scott Lewis, used it for some of the photos featured on his web site. Light painting is done using flashlights or other portable light sources to illuminate the subject and/or background of a photo. Typically it is done with a long exposure so that you have time to render the scene.

The light can either be used to illuminate the foreground subject, or it can be used to create the background by pointing the light towards the camera. Using colored gels on the lights provides you with a full palette of saturated colors to work with.

Over the last year I’ve been working with Katy in the studio, teaching her about photographic lighting. Last night we decided to play with light painting. I set up my Nikon D80 so it was tethered to my MacBook and using the free Sofortbild software we were able to instantly review each photo. The great fun of light painting is that you don’t know what you’re going to get. A digital camera makes light painting a lot more fun and certainly less expensive as there are a lot of throwaways. I can’t imagine trying the technique using film.

The anticipation of seeing each new creation show up on the screen was the best part of the process. Here are 5 of my favorites.

Pirating Under a Full Moon

Picture of the Month – September 2009

Because this month’s picture was created, rather than just photographed, I thought it would be fun to show you how it came together. I’ve done several pirate photoshoots both in the studio and on location, but this time I wanted to create something more theatrical.

The pirate was played by Michelle, who I met last year photographing her for her senior portrait. I shot Michelle against a blue backdrop that I new would be close to the shade of blue I’d want for the night sky. This made it easier to drop out the background and not leave a tell-tale fringe. The hardest part of lighting Michelle was getting the right balance of backlighting so that it would look like the moonlight was shining through her blouse. The color temperature for this shot was cooled way down to give it a blueish cast so it would look like a night shot.

A few days later I photographed the pirate ship against the same blue background. The pirate ship is a model that Andy Doerner built for Lori and I in 1999. It was lit in the studio to make it appear as the light was coming from the moon. I used one snooted strobe up on a boom to approximate the moonlight.

The sky and ocean came from a shot I took on a vacation in Mexico. Different treatments were done to both the sky and the water to get it to look like night. It seemed a waste to use such a colorful sunset, but I liked the highlights on the water and the clouds in the sky.

The moon was added from a public domain image shot by NASA. For the finishing touch, the stars in the sky were added on a separate layer in Photoshop.

Pneumatocyst

Last weekend while hiking to Punta Gorda I came across a lot of kelp that had washed up on the beach. Some fresh, and some that had obviously dried in the sun for awhile. Kelp has gas filled floats, called pneumatocysts, that help it reach the surface of the ocean to collect more energy from the sun.

We were finding lots of these floats on the beach, some the size of eyeballs, and others the size of softballs. One dried float caught my interest because of the little pigtail still attached. I put it in my camera bag thinking it might be fun to play with it in the studio.

So today I experimented with it to see what I could come up with. Depending on how it was lit, it either looked like a gruesome eyeball that had been plucked from some unwilling victim or some weird distant planet.

My favorite is in the planetary mode. It was photographed by lighting it from behind with an LED flashlight, and then from the front with two strobes, one with a cyan gel.