Digital photography has radically changed photography forever, not by continually offering more and more pixels, but rather by providing greater flexibility and control over the entire imaging process. With the entire imaging process becoming increasingly important, photographers need to adapt and learn the finer points of image making - blending both art and science. From the initial conceptualization of the image to a final image ready for viewing and printing, the creative imaging process goes through three distinct and equally important stages, namely:

• Imaging Stage I: Seeing - from concept to capture
• Imaging Stage II: Processing - from capture to image
• Imaging Stage III: Enhancing - from image to art.

Objective

This article is targeted at photographers and has for aim to take the reader through the identification and definition of image parameters that could and should be adjusted, tweaked, fine-tuned, customized, corrected, and optimized during the Processing stage.

The full Workflow of how each image parameter can and should be adjusted and optimized will be discussed in an upcoming series of articles by the author.

After having read this article, the reader should understand all image parameters and feel that his or her images will benefit from the optimization of each parameter discussed in the article. The reader will conclude that processing a digital image goes much further than just adjusting color and exposure.

Introduction

When left on "automatic", most consumer and professional digital cameras will only adjust exposure and color balance (although other parameters can sometimes be manually customized).

What does this mean to you, the photographer, and to your images?

This means that in an "average" photographic situation (meaning: a healthy and even amount of light from a single light source type and color, little difference between shadows and highlights, and a generous sampling of various colors), the processor of the digital camera will adjust the overall color balance as well as capture the correct amount of light (exposure, aperture, and ISO).

Although results can vary from extraordinary (with the more "average" situations - where the processor has a broader array of data points to make decisions) to mediocre (with the more "challenging" conditions like a monochrome scene with mixed lighting - where the processor has less valuable information to make decisions), most images will appear satisfactory - until one sees them processed skillfully.

To help the reader understand the image parameters that are presented, they have been grouped in the categories below and presented in the order in which they should be evaluated.

• Color
• Density
• Details
• Optics

Color

White Balance - White balancing is simply defining and determining what will be rendered as white or neutral grey (ergo also defining all other colors relative to white). This is done by adapting the sensor's reaction to the light source's color and type. Various light sources and illuminant types emit a different color spectrum (typically meaning bluer or yellower images - also referred to as cooler or warmer colors). With film, the options were mostly limited to daylight or tungsten - the small variations being corrected at the lab when making prints.

Digital cameras are more sensitive to different spectrums and have many more options such as daylight, cloudy, shade, tungsten, and neon to adapt to the illuminant. Most digital cameras also boast an automatic setting which evaluates and weighs the sensor(s)' information to select the right white balance as well as a custom setting where the photographer can provide the camera with a neutral reference point.

The real challenge arises in mixed lighting conditions (such as with on-camera flash indoors) where averaging the information may just mean an "average image" since the different light sources will be rendered differently within the same image. Precisely determining the white balance in these situations will require artistic and technical judgment as to which illuminant in the image (or combinations of) should be selected for the white balance calibration.

Color Profile - The camera color profile specifies the relationships between all of the colors so that once the correct white balance is achieved, all of the other colors [under the same illuminant] will be rendered accurately. Accurately rendering all colors in the rainbow is as important as achieving the right white balance. Color profiles can be specific to a software or even custom-made to your own specific camera body to account for sample-to-sample variations. To the extreme, lens-specific color profiles can be developed to account for the tonal signature of each lens design.

Saturation - With the right white balance and an accurate camera profile, all colors will be depicted faithfully and with the right saturation (color "intensity"). However, photographers can make an artistic decision to over saturate or desaturate all or only a specific color range(s) to achieve the visual or emotional impact desired.

Density

Exposure Compensation - Exposure compensation allows the photographer - typically within ±2-3 EV - to adjust the image "brightness" to compensate for any over or under exposure that would have happened during capture. However, there are costs to any adjustment. Decreasing the exposure of an overexposed image will most likely result in the loss of details in the highlights while increasing the exposure of an underexposed image will increase noise and provide less detailed color information.

Contrast - Contrast linearly changes the difference between the darkest shadow and brightest highlight (as well as any points in between). It should be noted that contrast is also linked to saturation - the higher the contrast, the higher the color satuation. Contrast can be used to restore colors and depth perception to images captured under overcast conditions. On the other hand, lowering contrast can be used to re-establish more natural colors in most consumer digital camera that tend to over saturate colors and exagerate contrast. Again, art and science blends at this stage as photographers must weigh between accuracy and beauty.

Levels - While contrast changes levels by resetting the black and white points symmetrically, levels take contrast one step further by not only allowing black and white points to change independently, but also by allowing the mid-point to shift the overall balance of the image from light to dark.

Curves - Curves take contrast and levels to an entirely new level. Curves can do what contrast and levels do - but in a much more sophisticated, precise, and flexible way. In addition to moving the white and black points, curves also "push around" the pixels along any given path.

Highlights and Shadow Recovery - Highlights and shadow recovery extends the perceived dynamic range by enhancing details in the most extreme zones of the image through a sophisticated localized curve and contrast algorithm. This is extremely useful when extracting details such as in a wedding dress and gathering details in a dark tuxedo. Used in combination with curves, highlights and shadow recovery are extremely powerful tools.

Details

Noise Reduction - Noise is the digital equivalent of film grain. To increase ISO, digital cameras amplify the voltage from the sensor, thereby "picking up" background noise (a little like the noise on audio cassettes). With higher ISO sensitivity comes higher noise (the equivalent of film grain). Noise is also affected by any density or color parameter modification. However, managed with appropriate capture techniques and with the proper tools and settings, noise can be drastically reduced during the Processing phase. Noise reduction is definitely an art that requires a lot of practice to obtain optimum results.

Sharpening - Sharpening works by increasing and exaggerating existing contrast to very small areas and edges. Because the human eye is very sensitive to contrast, the exaggerated edge contrast increases the perceived sharpness of the image. In addition, when used in the right proportion, sharpening can also restore the blur added by an anti-aliasing filter installed in most digital cameras to prevent moiré. Sharpening settings are dependent on the image content, camera, lens, output size, and ISO. Because sharpening may increase noise, it can also be applied selectively to portions of the image.

Local Contrast - Local contrast brings out medium size details by increasing contrast between a pixel and its surrounding area. By varying the size of the surrounding area, details of various sizes can be brought out.

Demosaicing - Photosites (commonly referred to as "pixels") are color blind. To make them see in color, a color filter is placed over each individual pixel (the most common type being the Bayer Matrix). Because each pixel only sees one color (typically red, green, or blue), the actual color is interpolated from an array of pixels. This image reconstruction process is called demosaicing (or Debayering in the case of a Bayer Matrix filter). Demosaicing algorithms are not yet a commodity so all software products have their own algorithm - some better than others. Selecting the right algorithm for the task is critical to overall image quality.

Bit Depth - Bit depth is the often forgotten variable. Increased bit depth offers more shades and colors so that information can be "pushed around" more precisely through the various adjustments and manipulations aforementioned. Selecting the highest possible bit depth (typically by selecting the right file format) gives the photographer the most amount of information, latitude, and flexibility. Quite often, photographers will debate between RAW vs. JPG. Unfortunately, the discussion is centered around which is "sharper". RAW files have the major advantage of containing much more bit depth as well as providing the opportunity to select the conversion software - and hence a demosaicing algorithm of higher quality with forward compatibility.

Optics

Distortion - Even the most expensive lenses suffer from distortion: a deformation of the image. This deformation is typically more prominent with wide angle lenses and exhibited as barrel (outward) or pincushion (inward) distortion - sometimes even both in zoom lenses. Fortunately, it is possible to re-map the pixels and eliminate any trace of distortion.

Vignetting - Vignetting is the drop in lens light transmission from the center to the corner of the image and more prominent at wider apertures. Although vignetting can be corrected or alleviated in most situations, it may not be possible without image degradation - the right capture and processing techniques are key. A new type of vignetting is also challenging camera manufacturers: sensor vignetting.

Chromatic Aberrations - Chromatic aberrations are born from trying to correct and "bend" rays of different wavelengths to focus on a single point. It is most visible at high contrast edges such as the edge of a building in bright daylight. Fortunately, it can be drastically reduced and even eliminated in some cases.

Conclusion

The Processing stage of the imaging process requires considerable attention to the color, detail, density, and optics parameters (as well as their interactions with each other).

Photographers will have to go the distance to learn how each parameter affects image quality (as well as other parameters) and follow a proper workflow in order to get extraordinary results.

Examples: Before and After