Monday, September 20, 2010

What Happens After Adobe Photoshop CS5's HDRPro?

I’ve had several inquiries lately about “how” I get the bright colors you see in my images and “why” I keep talking about using an individual Hue/Saturation Adjustment Layer for each color (Red, Yellow, Green, Cyan, Blue, and Magenta).  Along with today’s image I’ve included the Layers Panel that goes along with it.  Clicking on the image of the Layers Panel will produce an enlarged image of it and may make following along easier.  First, a little explanation of how the image got into the state it’s in.  It did start out as a three shot HDR Pro image.  The EV (Exposure Value) settings were -.3, -2.3 and +1.67, so it’s a three shot bracket at two stop intervals.  My “normal” EV setting is -0.3 stops, so that way you see what might look like strange numbers rather than 0, -2 and +2.  The reason for the “normal” being one third stop to the minus side is just to provide a denser initial shot.  The Layers Panel shown is only what was done after the HDR Pro work was completed.  The panel only shows the “finishing” of the image to its current state.  Of the fifteen steps in the panel you’ll see that eleven are Hue/Saturation Adjustment Layers.  Wednesday I show a Layers Panel that considerably calmer, but today’s is a little more on the extreme side.  There’s three Red Hue/Saturation Adjustment Layers, one Yellow, two Green, two Cyan, and one each of Blue and Magenta.  One of the first things you might notice is that every Mask that goes along with the adjustment layers, except one, are very different looking.  No two are alike.  The “how” and the “why” questions about the colors in my images both relate to my use of the Hue/Saturation Adjustment Layers.  Take a look at the explanation in the “read more”.

First, a quick statement about my use of the Hue/Saturation Adjustment Layers. It might as well be called the Saturation Adjustment Layer in the way I use it. I only use the saturation slider in the normal course of my work. I’ll run the slider up to 100% and then use Shift/Down Arrow to bring it back to the level that looks good to me. I use the Shift/Down Arrow to move the slider in increments of 10% rather than increments of 1%. The difference between 90% and 80% may be a visible change. The difference between 85% and 84 % won’t be discernible.

Let’s answer the “why” I use individual Hue/Saturation Adjustment Layers question. It should be pretty evident from looking at the image of the panel. Different colors need different masks. Sometimes the same color needs different masks in different areas. The first Red H/S AL concerns only the pillows on the couch in the windows. They need a more intense red (100%) than any other portion of the image. The second section of the image is the façade of the buildings. That’s 40%. The face and arms of the fellow sitting on the steps and the Open sign above him needed to be blocked. The final red adjustment was the sidewalk and the stairs going up into the building (70%). The Greens are also two separate layers. The first is a basic global adjustment (60%), but the granite steps and the two small, potted trees have been held back. The second green Adjustment Layer (80%) affects only the window box. And so it goes. Each image is looked at as its individual pieces. Does the pillow need more red than the façade? Does the window box need more green than the rest of the image? And on and on.

The Master Hue/Saturation option is just about useless. I can’t think of any image I’ve made in years where all colors needed the same amount of saturation boost. Using one Adjustment Layer and going through each color on that AL doesn’t work either. You’d be left with one mask, and as can be seen in today’s image, no one mask can could possibly mask each color effectively.

Once all the saturation levels were set for the individual colors a composite of the image was put up on its own layer (Crtl/Alt/Shift/E). This was used for dodging and burning selected areas of the image. For today’s image that was a straight Burn (O) of the basket hanging on the railing and an alternating Dodge and Burn of the folds of the man’s pants and sweater. Burn the dark areas and Dodge the light areas that fall right next to the dark areas to increase texture. The bottom of the roll of the sofa in the window was also Burned in a little to give shape to the couch.

The final steps were to make another composite of the image (Ctrl/Alt/Shift/E) and then a copy of that layer. The first composite layer was used to sharpen the image using a High Pass Filter after creating a Smart Object out of the layer. The copy (unsharpened) was used to create a vignette. The Rectangular Marquee Tool (M) was used to make a selection equally around the image, then feathered to round the corners. Tapping the Delete Kay makes a hole with a feathered edge. Change the Blend Mode to Multiply to darken the edges and corners of the image and use the Opacity Slider to adjust to taste.