Washtenaw Community College Don Werthmann School of Digital Media Arts Photography GM300BB 973-3586 http://courses.wccnet.edu/~donw donw@wccnet.edu What is a Digital Image? Any digital image is composed of square picture elements, more commonly known as pixels, in every stage of its life. Pixels are storage bins of data generated by digital cameras and scanners; viewed in monitors and projectors; and then can be output to print. A pixel doesn't have any fixed width or height; its size changes relative to the device it is being used with. Each pixel is finer than a human hair, and is assigned a numerical value that can produce 256 shades of gray in Grayscale mode, and over 16 million colors in the RGB Color mode. Millions of pixels put together form a grid, and collectively, they form the illusion of a continuous-tone image a photograph. Pixels are the foundation of digital images. Binary System A form of mathematics invented by Gottfried Wilhelm von Leibniz (1646 1716), in 1679, became the basis of virtually all modern computers. With the Binary System, a computer can create all other numbers using just 0 s or 1 s. Number values are assigned to individual pixels which, when assembled into mosaic-like collections, convey visual information. Bit Short for binary digit, it is used with exponential notation that creates an electronic signal the computer interprets as information. The electronic signal controls a bit to tell it whether it is either 0,"off," or 1 "on." The bit is the smallest unit of digital information available. Byte Short for binary term, a byte is eight consecutive bits, such as 0 1 0 1 1 1 0 1. In a byte, each bit is a fixed value in geometric progression, meaning that each consecutive bit doubles in value compared to a bit that precedes it: 1, 2, 4, 8, 16, 32, 64, 128, etc... 0 1 0 1 1 1 0 1 1 2 4 8 16 32 64 128 This byte, 0 1 0 1 1 1 0 1, is the value 186 because 2+8+16+32+128=186. Why does 8-bits make a byte? Historically speaking, because of two factors: 1) the earliest form of computer technology was limited to sending only 8 bits at a time, and so it became a natural barrier for computer programmers to write code in sets of 8 bits. 2) it s the minimum number of bits that are required to represent a single character from a Western character set. A Pixel in Grayscale is a Byte (8-bit mode) To render the illusion of a continuous-tone image, each pixel must, at a minimum, be capable of functioning in an eight-bit mode. This also means that each pixel is a byte of information. This image is grayscale, or what s commonly referred to as a black-and-white photograph, so the pixels only render various levels of luminosity, without color. If 1+2+4+8+16+32+64+128=255, then each pixel can produce a maximum of 256 shades of gray, including 0. 0 is black, 128 is middle gray, and 255 is white. Another value in that range, such as 186, renders the pixel as a shade of gray value 186 is light gray.
Digital Photo Imaging PHO 127, page 2 of 5 Kilobyte K [1 thousand] A collection of 1024 bytes is called a kilobyte 1024 bytes = 1 Kilobyte. Because of the binary system s geometric progression: 1, 2, 4, 8, 16, 32, 64, 128, 256, 512, 1024 People frequently round down a Kilobyte to 1000, which is derived from the metric system. Megabyte MB [1,000 x 1,000 = 1 million bytes] 1024 Kilobytes = 1 Megabyte. Gigabyte GB 1024 Megabytes = 1 Gigabyte. Terrabyte TB 1024 Gigabytes = 1 Terrabyte. Strange but true 1024 TB = 1 Petabyte 1024 ZB = 1 Yottabyte 1024 PB = 1 Exabyte 1024 YB = 1 Brontobyte 1024 EB = 1 Zettabyte 1024 BB = 1 Geopbyte Random Access Memory RAM Every computer has RAM to simply operate and run applications and create documents. The computer copies and starts an application from the hard disk and allows you to create an image, a letter, or whatever. Warning: You must SAVE a document you're working on to a hard drive frequently. If you don't it is in danger of being lost due to a power failure, system/software application crash, or other event. Once the computer is off everything in the memory (RAM) is GONE and that's not an acronym. Cache Pronounced "cash," this is a part of the memory the computer uses to hold information about the most recent things done on your disk. Its capacity level can be adjusted. Megahertz MHz or Gigahertz GHz 1 MHz = One million cycles per second. 1 GHz = One Billion cycles per second. Used as a radio-frequency unit, but here, to describe how fast the computer can process commands. The higher the number, the faster the computer operates, and the more expensive it is. Central Processing Unit CPU This is the thinking portion of the computer. You'll hear computers described in terms of which CPU is installed. Internal Hard Drive Every computer has a hard disk. Consider this the filing cabinet where you store documents within folders. Computers are sold featuring the size of storage space, usually in Terrabytes. The bigger, the better, on both memory and hard disk space. External Hard Drive These usually have a smaller or similar capacity than a hard disk built into the computer, but allow you to store/transport files easily. Keep them away from extreme heat: like in your car at mid-day. Keep them away from magnetic sources a magnet will destroy the data. Don't store them near magnetic fields such as a telephone or stereo speakers, or other electronic devices. Keep them away from static electricity. Airport security X-ray machines will not harm them. Operating System OS Software that gives the computer its "look and functionality." It s continually being updated. Try to keep up to date with a recent version number, but read some software reviews first to make sure it s compatible with other hardware and software you already have. Complementary Metal Oxide Semiconductor [CMOS] This is the light sensitive sensor that captures light, processes it, and turns the energy intensity recorded into a digital signal that a computer understands.
Digital Photo Imaging PHO 127, page 3 of 5 Resolution Resolution determines the quality of an image s appearance across various viewing platforms. ppi refers to pixels per inch to produce an image on electronic devices [pixels on a monitor] dpi refers to dots per inch to produce an image on printing devices [dots of ink on paper] Image Size & Resolution The size of an image is controlled by its available quantity of pixels [width x height], and the total pixels are used to calculate its potential output dimensions. Resolution refers to the number of pixels per linear inch, which is determined while considering output intent [monitors vs. printers]. Resolution for monitors varies because hardware quality covers such a wide range use 72ppi. 240 or 300ppi is advisable with Epson printers. File Size [noted as Image size in Illustration] A digital file s size is calculated by the quantity of pixels available times the number of colors in the Color Mode photographers use RGB. The file size affects how many images can be stored on a hard drive or camera storage card. Higher resolution images make larger file sizes. Two images can have the same physical output dimensions [document size ~7x10 ], but the image with a lower resolution has fewer pixels, and is therefore a smaller file size. This digital capture provides over 6 million pixels to work with 2104 x 3000 = 6.3 million The Image Size Dialog determines available data that can be changed & optimized based on the output intent [monitor vs. printer]. 300 [ppi] x 7.013 [inches] = 2104 pixels 300 [ppi] x 10 [inches] = 3000 pixels 2104 x 3000 = 6,312,000 pixels [6 Mgpxl] The File Size is determined by the quantity of pixels available x [RGB] the number of color channels in the capture 6.3 million x 3 ~ 18 Mb the same image in CMYK is approximately 24 Mb. the same image in Grayscale is approximately 6 Mb. Document Extensions The proper way to name a document is to "tag" it with a native, three-letter extension. An example being a Photoshop document entitled my_image.psd my-image-dot-p-s-d. There is an extensive list of extension acronyms, and each provides a clue as to what kind of file it is. JPEG [.jpg] Joint Photographic Experts Group. A mathematically compressed file format that uses less disk storage space. The advantange of compression is quicker transmission and economy of size. The disadvantage is the loss of detail and color data, therefore it is known as a Lossy file format. This is a universal file format that most software applications are capable of opening. TIFF [.tif] Tagged Image File Format. This is also a universal file format, which many applications can open. There are optional compression rates are available to economize disk storage space. These mathematical compression algorithms used are not detrimental to the image s quality, and is therefore known as a Lossless file format. The format (owned by Adobe) is publicly documented which greatly increases the longevity of opening an image in the future. TIF is very robust because it supports Photoshop layers. A flattened version of the file is embedded within the file so that applications other than Photoshop can utilize them, and therefore, the file size is typically larger than a PSD of the same layered image. PSD [.psd] Photoshop Document. This is a unique [proprietary] file format that can be opened only by Adobe Photoshop or another application in their creative suite of software applications, such as Illustrator, Premiere, AfterEffects, InDesign, etc. Like TIFF, this is a Lossless file format.
Digital Photo Imaging PHO 127, page 4 of 5 Bit Depth This refers to the number of bits used to store information about each pixel. If the number of bits assigned to each pixel rises, then the computer can display more shades of gray. 8-bit is a common mode [bit-depth] to work in, but 16-bit has become more viable because of the increasing affordability of a CPU, RAM, and hard drives available to working professionals. Consecutive strings of bits form different combinations of numbers. For example, one-bit graphics are only capable of showing two things; black [0 off ] and white [1 on ]. In binary code, a two-bit pixel has four number combos 00 or 01 or 10 or 11; a three-bit pixel can look like 010; four-bit 0110; five bit 01010; six-bit 101010; seven-bit 0101010; eight-bit 10101010. Each increase in bit depth is factored by 2, since each digit can be either 0 or 1. A 3-bit pixel has 8 combinations 2 3, or 2 x2 x 2 = 8 levels of gray can be produced. A 4-bit pixel has 16 combinations 2 4, or 2 x 2 x 2 x 2 = 16 gray levels. An 8-bit pixel has 256 combinations 2 8, or 2 x 2 x 2 x 2 x 2 x 2 x 2 x 2 = 256 levels. 256 ways to arrange eight 0 s or 1 s: 00000000 = black; 11111111 = white. 1-bit image 2-bit image 3-bit image 4-bit image 8-bit image The pixels of a 1-bit image can only render an image with black or white. That s it. The result is extreme contrast, with absolutely no detail offered to render the subject matter. As bit-depth increases, more gradation of tone is possible and more detail can be offered to define the subject matter accurately to the human eye. 255 is a critical number to remember because the maximum numerical range for a digital image is 255 [inclusive of zero it's 256], as seen in a histogram [Levels Adjustment]. As you can see, greater bit depth leads to greater tonal range (shades of gray) per pixel, which provides greater image detail, and a more gradual transition of tone & color across the image. When colors are introduced to each pixel, it leads to more possible shades of color. Three colors per pixel red, green, and blue make 256 x 256 x 256 = 16,777,216 colors possible. 8-bit color is also known as 24-bit color. 16-bit color is also known as 48-bit color. 16-bit Pixel [16-bit mode] A 16-bit image has 65,536 combinations that s 2 16, or yes, 2 x 2, 16 times. In case you re wondering, 65,536 x 65,536 x 65,536 = 281,474,976,710,656 possible colors. 0 1 0 1 1 1 0 1 0 1 0 1 1 1 0 1 1 2 4 8 16 32 64 128 0 1 0 1 1 1 0 1 0 1 0 1 1 1 0 1, is the value 47802 because 2+8+16+32+128+512+2048+4096+8192+32,768=47802. If 1+2+4+8+16+32+64+128+256+512+1024+2048+4096+8192+16,384+32,768=65,535, then each pixel can produce a maximum of 65,536 shades of gray, including 0. 0 is black, 32768 is middle gray, and 65,535 is white. A pixel assigned with the value, 47802, renders the pixel as a shade of gray value 47802 is a mid-tone. Anyone up for 32-bit? 256 512 1024 2048 4096 8192 16,384 32,768
Levels [a.k.a. Histogram] This mountainous profile of information is a graph that interprets the distribution of tone throughout an image. There s no ideal shape to histogram data, as they appear differently image-to-image, but an ideal placement of data is skewed toward the right. There are 256 data points on the bottom of the graph 0 is black, 255 is white. The number of times each value occurs in the image is indicated by its frequency the higher the altitude, the more times the value is found in the image. A high solid line at the black point or the white point [when data hits the wall ] means there is most likely more information available from the capture or scan. The image is either under- or overexposed. The levels dialog [right] is powerful. This one tool can control three critical attributes of a photograph: density, contrast, and color. Digital Photo Imaging PHO 127, page 5 of 5 Channels Photoshop supports images in 8-bit and 16-bit modes per color channel red, green, and blue. Eight bits per channel, times three channels [red, green, blue] equals 24-bit color. [8 x 3 = 24] Click on the Channels tab in the Layers palette. Each color [RGB] is represented in grayscale. Data that is white [255] allows100% of the color in that channel show through. Data that is black [0] allows 0% of the color in that channel show through. 100% of all three channels [RGB] = pure white, whereas 0% of RGB = pure black. All three channels aligned pixel for pixel create the full image formation Master channel Grayscale is one channel and represents pixel luminosity [no color data in the pixels] Under the graph, find the small triangular-shaped sliders. The left slider establishes the black [0], the middle slider establishes the middle-gray [128], and the white slider establishes the white point [255]. Use the gray slider to adjust density or image brightness [value 1.0 represents gamma adjust]. Use the black and white sliders to adjust contrast. Use the gray eye-dropper to adjust color balance. Layers Layers are individual images stacked on top of one another, that work much like stacking individual sheets of clear acetate. Each layer has its own set of channels, and the top layer obscures the ones below it. If you see a gray and white checkerboard pattern appear in any portion of your image, on the monitor, then you are seeing transparent areas through all layers, down to the canvas. Adjustment Layers Layers Palette: (Half-moon icon) New Adjustment Layer... Several kinds of corrections can be made to an image, with the flexibility to revisit the changes you made and modify its settings or values. Similar changes can be made to a layer directly, but those values will reset to zero each time it is revisited. Color Balance Adjustment Layer If you perceive a color cast in the image, use its complementary color to neutralize it. Click on the respective Tone Balance radio button shadows, midtones, or highlights to affect color correction to each value range independently.