How a Pixel Gets its Color | Bayer Sensor | Digital ImageShow Video Details ↓ [silence] …Narrator: The light catching buckets on a sensor are not really pixels in the traditional sense even though we all refer to them as pixels. There are more accurately referred to as sensels or sensing elements, or sensels which capture light. Most cameras today use something called a Bayer filter system. A complete Bayer set is made up of four colored filtered sensels: Two green, staggered with one red and one blue. Think of a filter as a very small piece of colored film covering the light sensel. Its job is to prevent certain types of light from entering. So let me ask you this: what do you think happens when different colored light hits one of these colored filtered sensels? If a red filtered sensel is exposed to a bright blue or green light, the sensel cannot see it. However, red light can penetrate a red filter and the volume, or the brightness of that light, is recorded by the sensel. The camera's processor knows that if any light penetrated the red filtered sensel it can only be red. And therefore it calculates that pixel as red light. So the color calculation happens in the processor, not by the sensor itself. Blue filtered sensels will only capture blue lights and green would only captured green. For the rest of this lesson well use colored water to show you what the processor sees. Just remember it starts out as an intensity of light, not a color. The information of red, green and blue light which makes up an image is also referred to as channel information or channels. Now I imagine you have several questions. You know that when you magnify an image, you see several colors. Not just red, green or blue but you also see yellow and orange and aqua and purple and pink. So, you probably wondering what in the world is going on. For every pixel we see, in a digital image, nine sensels are involved in the calculation. One central sensel, which will be the position of the target pixel, as well as the surrounding eight sensels which will contribute their information to the processor's calculation. Now just for fun say central sensel nine times as fast as you can. That should keep you busy for a little while. Now let's simplify this a little bit and instead look at three sensels, a red, green and blue with the middle sensel acting as the target pixel area. Now we already know that if the sensels are exposed to red light, the target pixel will be red. If the sensels are exposed to blue light, the target pixel will be blue. And probably we already guessed that if sensels are exposed to green light the target pixel will be green. OK now those were pretty easy. What about yellow light? Well, yellow light is made up of both red and green light. The red and green sensels both capture light and the processor knows that when this happens the target pixel is yellow. Aqua light is made up of both green and blue light. Purple light or magenta is made up of both red and blue light. If the sensels don't catch any light, the pixel will be black. If the sensels are completely full, the pixel will be white. If the sensels have any equal amount in between those two it would be a shade of gray. In fact, different amounts of red, green and blue information can make just about any color of light. We know that in JPEG images, which use eight bit depth, there are total of two hundred and fifty six shades, or intensities of light. If you were to calculate the total number of possible colors it would be two hundred and fifty six red intensities times two hundred and fifty six blue intensities times two hundred and fifty six green intensities, which equal about seventeen million possible colors. That's amazing. Now you can also see this in action using the color picker in Photoshop. The numeric values you see at RGB stands for the amount of red, green and blue information used in calculating that specific color. Do you notice how these values stay between zero and two hundred and fifty five? Every time we pick a color using this swatch, we can see how much of each is used in the calculation. Black has all zeros. White has all two hundred fifty fives and grays are all equal values. If we were to look at any other color we can see how much each channel contributes to that particular color. Now we don't think about it much when we working in Photoshop but we're using it's tools to make adjustments, what we really doing is changing the pixel's appearance by reassigning the input information. We're changing these numbers. That's what Photoshop does when we make adjustments and I think that's a pretty cool concept to know. I'd probably say ninety to ninety five percent of all photographers and Photoshop users really don't understand that this is happening in Photoshop. So hey, you know that's a good little conversation starter. Do you know how pixels get its color? OK, you can talk about that at, you know, your next little photo, you know, meeting or whatever you guys do. The take home message for this lesson is that every pixel gets it color from gathering red, green and blue information from several sensels. And it's the processor that's able to piece this information together and decipher the proper color for a pixel. It's pretty much amazing if you ask me. If you found this video helpful, you may be interested in my new DVD Photoshop crash course. I'll not only teach you the most important tool in Photoshop, I'll teach you how to think in Photoshop. It can be ordered from the following link. [silence] … |