In photomechanical imaging, the characteristics of the image are self-evident: it includes tonal and color information, along with a grain or dot structure. This corresponds to the most basic digital images, which were originally just a rectangle full of colored dots representing tone, color, and resolution.
As technology evolved, digital images have become far more complex. At first, this was playing catch-up with physical images. For more than a decade now, digital images have included components that extend our understanding of exactly what an image is. Here are some examples of that evolution.
Color management
The numbers that are used to describe colors could be calculated in many different ways, Color management is a system to describe exactly what is meant by a numerical value. It enables standardization of color rendering across devices that may have very different color signatures (or very different models of rendering color).
Compression
It’s always been important to shrink digital images down as small as possible for any given use. By standardizing ways to compress and expand image data, storage, and transmission bandwidth are conserved.
Multiple images
For several decades, it’s been possible to place multiple images into a single file to create a new synthesized image. This allows images to be composited and also to be adjusted, and re-edited.
Transparency
While most digital images are rectangles, adding the capability to make some pixels transparent allows an image to take on any conceivable shape.
Adjustment instructions
Adding color and tonal adjustment instructions to an image allows the “photo finishing” to be attached to the image itself and for it to be readjusted non-destructively. It also allows multiple interpretations of the same image to be “part” of a single original image.
Masks and alpha channels
Digital images can contain other images – often a monochromatic image that can be used to mask parts of the primary image. This can be used to create a transparency of the image, or selectively apply adjustment instructions to regions of an image, or blend multiple images together.
Curved and spherical models
Digital images may also be created as inherently non-flat photos that use curved or spherical geometry. Spherical images can be viewed in a VR environment or on a flat display that is designed to compensate for the curvature of the image.
Motion and audio
Adding multiple sequential images to an image can create a moving image. Hybrid image files have been around for decades. Likewise, it’s been possible to attach an audio file to an image for a long time.
Camera raw images
Digital cameras created a new type of digital image – one that preserved camera-native information for flexible reprocessing at any point in the future.
At this point, the nature of digital images changed radically. Entirely new image editing methods were needed, as were ways to store this unprocessed data properly.
Depth information
Mobile phones and other computational cameras are now creating depth information as an inherent part of the image itself. This is useful for all kinds of purposes, including synthetic depth of field, Augmented Reality features, facial recognition, and other visual analysis.
Application-specific usage
Mobile imaging, raw photography, and cloud workflow have created a need for even more complex image objects. This includes the need for proxy images and application-specific rendering, all as part of a single object.
All of these above items can be integral parts of the image itself since they inform the nature, understanding, and use of the images. Many images will continue to be simple rectangles of colored dots, but an increasing number will be more complex digital-native objects. In the next post, we will take a look at the ways that formats have evolved along with the images themselves.