|
|
|
|
|
||
|
||
|
This short basics post will prime you to understand how colors are specified in digital files. In the reproduction market, of which Reed Art & Imaging is a part of, we use digitally driven devices to make faithful reproductions of original art, photographic captures and digital graphic designs. To accomplish this task with any hopes of repeatable accuracy, there must exist a standard system by which colors can be recorded, transferred, translated and output. These standards exist in theoretical color models. These models are a virtual shape, such as a box, sphere. polygon or other shape that if it were real, would contain every color visible to the human eye.
Because the model is represented by a shape, they are referred to color “spaces”, for the space the object would occupy in the theoretical environment of all colors – visible and invisible. The graphic above is an example of a space that uses Red, Green, and Blue to yield the final color we want to create.Colors come to our eyes in two ways – or transmitted from a light source or reflected off of a surface.
RGB is called the “primary” space and it’s numerical system can be equated to the brightness values of transmitted light – or how intense the Red light, Green light, and Blue light are shining. As the numeric value increases, the lights get brighter and the closer to white they become. More on that in a bit.
In a CMYK model (the secondary space) we are representing pigments that absorb light. So as the number increases in their scale, the more light is absorbed. So with CMYK, the higher the number, the darker the color appears – exactly opposite of RGB.
In either space, the ratio of how the colors are blended determines the color, while numeric values contribute to how bright or dark it is.
For simplicity, the rest of this article will use only one color model. I’ll use the RGB model for these examples because it’s the model that our clients use and best supports high-end reproduction digital printing.
Color is usually expressed in human terms by it’s
In the data driven world, it’s expressed as a recipe of the colors required to build its final value, saturation and hue. Image and graphics applications usually use the standard scale of 0-255 ( what is called 8-bit color) to represent the amount of each color present, with 0 being none and 255 being maximum. Dark colors being closer to 0 and light colors being closer to 255. Equal amounts of each color create neutral hues ( grays ) and as the numbers increase from 0 to 255 the value moves from black to white.
These numbers from 0 to 255 are called “Levels” and in our examples fall into a model of 256 levels – with zero being included as a level. In an RGB color space, each color is built using various levels, or recipes, of Red, Green and Blue. Dark Red has a different recipe than Light Red, and the recipes are different for a saturated versus less saturated red.
As you can see in the first example above, a fully saturated hue has 255 of it’s requisite colors and none of the other colors. As the color desaturates, it gains some of the other colors; it’s moving closer to a neutral gray. In the second example we can see that the Darker Red contains none of the other colors, but the Red number is dropping closer to zero; thus making it “blacker.” This darker red is as saturated as it can get at this present value.
A critical point to understand is that in an RGB or CMYK file, color and density are inter-connected. Meaning that any change you make to color data will result in changes to density and visa-versa.
The other primary colors are built in the same way, like this:
The secondary colors are built from equal amounts of two of the three colors:
These secondary colors are thought to be the “opposite” colors to those in the previous example. You will notice their recipes are directly inverse. Red is R255 G0 B0 and Cyan is R0 G255 B255. They are opposites because when the two colors are combined, they cancel each other out and make gray. Equal parts of Red and Cyan make gray, same goes for Green with Magenta, and Blue with Yellow.
Intermediate colors such as Orange, Brown, Purple, Daisy Yellow, Lemon Yellow etc. are built by using various values of the three colors where at least one of the colors is greater than 0 and less than 255:
This 8-bit model, using it’s 256 level per color channel architecture allows for approx 16.7 million variants of color and density. (256 x 256 x 256 = 16,777,216).
Other bit-depths exist that extend the number of available colors; the concepts are the same, but the numbers differ.
For example: 12-bit color – the depth that most digital cameras record in raw format, has 1,728 levels per color channel (instead of 256) with a total number of 5,159,780,352 available colors, much higher than present technology can reproduce in a print or display. The commonly used 16-bit depth has 4,096 levels per color channel with a total number of 68,719,476,736 available colors – yes that’s 68.7 Billion! While some professional pigment printers and their RIPs can support a 16-bit file, getting the subtleties from that many colors on paper and dots via a limiting 8 to 12 different ink colors is still problematic.
If you have questions, post them in the comments below. If you want to see how this all ties together with Photoshop channels, stay tuned, that’s next!
Color channels are often thought to be the exclusive realm of mystics and Photoshop gurus. If you are willing to dedicate a few minutes of time to learning, I’ll take the mystery out of channels, and give you the power to improve your workflow and your end results.
The colors we see on our monitors and in print are created by combining specific amounts of either Red, Green, and Blue, or Cyan, Magenta, Yellow, and Black with the result being a new intermediate color. Since the majority of our readers are using the RGB model, I’ll stick with that for our examples. If I get requests in the comments below, I’ll add a section explaining CMYK.
Most users of image editing applications like Photoshop or Gimp, as well as users of other graphic design applications are familiar with, or have heard mention of the 256 levels used to define color and density. Most often these levels are represented by numeric values from 0 to 255 with zero being one of the levels. In the RGB model, these levels can be equated with visual light, zero being no light, or pure black and 255 being maximum light – pure white.
When we build colors in the 8-bit RGB model, we are using 256 levels of Red, Green, and Blue in various combinations called a “build”. You can think of the color-build as a recipe for that specific color.
Collectively the color channels are nothing more than a representation of those recipes. And when the recipes for all the pixels are put together in the right order, we have our color image. Viewing our color channels is effectively changing the way your cook-book is organized. So rather than finding the recipe for the pixels on one page of your cook-book, your color cook-book has three pages, one each for Red, Green, and Blue. The Red page tells you how much red to use and where, the same goes for the Green page and the Blue page. So in our example above if we assume that each colored square represents 1 pixel, the Red page would tell us the first pixel would have 255 red, the second pixel would have 68 red and the third pixel has 126 red. The Green page would read: 1=128 and 2=68 and 3=0 and so on for the Blue page.
Photoshop shows us these channels in a way that our minds can easily process: as images. We can grasp the concept of images much easier than looking at the potentially millions to billions of numbers required for single image. Photoshop’s default is to show you these images as various shades of gray (256 possible shades to be exact). Here is what our example above looks like as color channels:
Where the build calls for zero of a color, that channel represents the area as black. Where it calls for all of that color, it is represented in the channel view as white. All intermediate values show up as the appropriate shade of gray.
This image is pretty much straight out of a raw conversion. The file has been optimized in the conversion to make sure that none of the channels contain either pure black or pure white. This is to mimic the way the eye naturally sees. We’ll compensate for its somewhat flat appearance when we show you how to optimize your files without damaging your color fidelity.
Here is the view of the red channel, remember lighter areas indicate more red, darker indicates less:
Here are the Green and Blue channels, you can click them for larger viewing:
Notice that the lighter areas of the scene show as lighter in all three channels, and the darker areas of the image are darker in all three channels. You can also see that the areas of the image that are green show as brighter in the green channel in relation to the other two.
Also, all three channels have complete detail from shadows to highlights, nothing is lost. This is critical for full color fidelity. This full range of detail is essentially how channels should be. When channels look muddy or if there is “clipping” to full black or full white, there is a loss of color fidelity. I use channel views regularly to examine the state of a file’s “health”. If the file’s channels are not right, then I know right away I can’t generate the best possible print.
It is key to understand that in an RGB color space, a channel is both color and density information. Any change that you do to a channel will affect the color, saturation and density of your file. If you increase any value in a color channel, let’s say moving the red value of an area from 180 to 185, the resulting color will be more red and lighter.
See, no mystics required.
Reach out in the comments below with questions and comments.
Okay, I get it, right now your likely asking: “Why all this search engine stuff, I thought this was about the picture I use on-ine?” Well – it is, and all of the search engine stuff is a just a beginning to explain why your head-shot quality is important to your business survival. Gone are the days that your head-shot was just so humans could recognize you. Google wants to make sure that any good content you create is appropriately credited to you, and someone else’s junk content is associated with them and NOT you. According to video interview featuring Mark Traphagen of Stone Temple Consulting: part of how Google identifies your content is through facial recognition of your head-shot, and partly through other technical means – and the latter can lead to mistaken Identity. The better the headshot – the better the chances they will know it’s you while no head-shot at all could result in failure. This is not to say that your mug is the only way they know it’s you, but it ups the odds for accuracy. To keep things working smoothly, it’s recommended that you use the same avatar across all your social media accounts.
The first step in your customer’s buying cycle is to recognize they have a need. The second step is to learn how to meet that need – and preferably through doing business with you. Almost all the information gathering about a business, service or product starts on the web, and now, the great majority of it begins with a search engine. The search results delivered to the user is moving away from how well our website is optimized for keywords and towards deep dependency on what really matters to people: your reputation on the internet and what the user is actually looking for. How you present yourself online – i.e. the good and the bad of what you post, is added or subtracted to what others are saying about you in conversation and online reviews, to generate a reputation score. As David Amerland, author of the book Semantic Search so perfectly stated in an online broadcast via Google+: It’s a shame that it took software to make us behave online, but that’s the truth of what is happening.
And behave you should – assuming you care about the future success of your business. Google no longer sees your business as separate from you, but rather your business AS you. Post a ton of negativity on the web and your business can suffer as a result. All things that lead to you, or anyone that Google knows for sure contributes to your business website, blog, social media, etc. may be factored into your business’ online reputation scores. And you certainly don’t want a mistaken identity tarnishing your hard-earned reputation.
Increasingly your personal brand will factor into your business or your employ-ability. Prospective customers and employers are looking to social media and the web to learn about you before giving the green light to proceeding any further with you. It’s really just a matter of time before software emerges that can give the user a reasonably accurate estimate of your reputation score – along with those with whom you are competing. I estimate that soon your personal brand will carry as much weight or more than your résumé when decision time comes. If you change companies, your score goes with you, while your companies score remains with them. Your score could some day soon become a marketable benefit to hiring you.
Recognition for a company comes in the way of their logo, or a distinct appearance of their products. For you, this recognition comes in the way of your face. So in the very same way that a business keeps the same “visual identity” across all of it’s marketing, so should you. And just as well designed identity-package is important to a brand to build buyer trust, so is a well done head-shot important for your personal brand.
Presently the generally accepted suggestions are this:
Some pro’s will offer a discount to shoot your entire office or team in the same session. If you are a solo-preneur, call a pro and find out how many people you need to get a discounted rate and then call your friends, associates, or work out some other way of getting the minimum numbers you need. I suppose you could have your Ol’ Uncle Joe take the shot with your cell phone, but let’s be real, we can ALL tell the difference between a shot done by a “friend” and one done by a seasoned pro. You can bet your prospective customer can too. Purchasing only happens when the buyer has confidence. A not-so-great shot doesn’t exactly scream “You can trust me to do my best for you”.