Create High Quality Pictures Using High Dynamic Range Rendering

Haven't you noticed that there are some times, when you take a picture, there is no way you can make the illumination be right? If your background looks fine, your foreground doesn't and viceversa . This happens because cameras can't reproduce scenes with high contrast the way our eyes can perceive them.The dynamic range is the measurable difference between the brightest highlight and the darkest value. This range is quite limited in sensors, that's why when we take a picture we must decide which exposure level to use, depending if we want to preserve the highlights or the shadows. In order to get a high quality picture, several techniques can be applied, one of them is HDRR , which stands for High Dynamic Range Rendering. This is how the rendering of computer graphics by using lighting calculations in a wider dynamic range is called. This technique imitates the human eye which adapts to all the different luminosities of a same scene by opening and closing the pupil while all the information gets together. For this reason, we can see zones with different illumination that photographic cameras can't show.

The HDRR concept has been around for a long time, but it wasn't until a couple of years ago that was highly developed and used. We can see it today improving photographs and even in video games such as Half-Life 2. It's mainly used to improve the representation of dark and bright areas in an image, and besides, greatly augments the contrast by making bright things look brighter, and dark things darker. Additionally, flat scenes look much better by using this technology, because light and dark can be represented with a greater variety of levels.

However, we also face some limitations when we want to use HDRR. Most displays can't represent HDRR images because the values exceed what they display can show. In order to fight this limitations, two techniques are used: Tone Mapping and Light Blooming. Tone mapping is a technique which maps the colors so they can be shown in an output device and stills preserve the image and the colour to appreciate it. Light Blooming is a graphic shader effect that is mainly used in video games. It's used when we need more brightness than the one our output device can show. To do so, Light Blooming expands a source of light by multiplying the image, blurring it and then redrawing it over the original one, that way, if there is a brighter image that needs to be shown, the illusion is created by making the scene softer.

Now that we know what a HDRR is, we'll talk about how we can create one. The basic thing we need in order to create a High Dynamic Range Rendering are three pictures with different exposures. You can work with more and get better results, but three will work. In order to get different exposures you can either change the exposure compensation of your camera or change the shutter speed. When you take the pictures make sure you use a tripod, that way you'll make sure your shooting the same angle, and you can overlap your pictures later (if your camera has the bracketing option use it too). Using ISO 100 is highly recommended since the less sensibility the more exposure time (for the same reason be sure to get a tripod, or else your picture will look blurry!)

When we have our pictures, we can easily create our HDRR images through our favourite image editor, like Photoshop.

In order to create a HDR image using Photoshop, we have to go to the File menu, Automate, Merge to HDR. We select our pictures and make sure we select the "Attempt to Automatically Align Source Images" option. We should get something like this:
http://www.iconshock.com/blog/wp-content/uploads/2008/06/1st-300x185.jpg

Photoshop automatically determines the White Point (Lightest tone of the image), but you can modify it if your lighting tends to be "burned". If you choose to work on 32 bit/Channel image, this is pretty much where it ends, but if you want to work with 16 bit/Channel images you will get some more features to improve your photo or create nice effects.

Exposure and Gamma allows you to adjust the exposure (brightness) and gamma (contrast). Highlight Compression applies a tonal curve which reduces highlight contrast in order to restore contrast in the rest of the image. Equalize Histogram is a method which spreads the histogram peaks, making our picture homogeneous. Finally, Local Adaptation is the most flexible way to modify the HDR picture, since it does local contrast enhancement; the tonal curve can be adjusted to get better results.

Examples:
http://www.iconshock.com/blog/wp-content/uploads/2008/06/uno-300x202.jpg
http://www.iconshock.com/blog/wp-content/uploads/2008/06/dos-300x201.jpg
http://www.iconshock.com/blog/wp-content/uploads/2008/06/tres-300x201.jpg

If you want to create a HDRR image of a moving element you can set up your camera in order to take pictures in a RAW format. A RAW format is a file type proper from every camera fabricator that saves the image just like the camera sensors perceived it. You'll get bigger files but you can change from one single picture the values of exposure of the picture.

But, if with one single RAW image you can create your HDRR image, why bother to take several pictures in a JPG format? Well, since RAW images depend on the camera's sensor also have a limited range, and, even if it's wider than JPG's range, doesn't give us as much quality as the one we can get by taking several JPG pictures and overlapping them. Bright objects, strong textures and color differences will look great in the final HDRR picture.