Digital Photography 101 - Part 4

Photography 101 – Part II

The Digital System

The Digital system – Digital photography introduced a system that never before existed in the world of film photography with many advantages such as practically endless photo number, immediate feedback on the photo taken, and lately even higher quality than film based cameras. in this part we will walk through the digital system and understand its parts.

a. The Sensor – That's actually our "film". Its job is to capture light and convert it from photons (light particles) into electrons by photoelectric cells. Each cell in the matrix transfers electrons depending on the amount of light that enters it. A proper analogy is buckets filling up with rain water (buckets being the sensor and rain is photons), each cell has a maximum capacity of "water" and if it fills up the pixel is white and can also affect the cells around it, and this effect is called "blooming". In modern sensor we can find a "drain" system that helps prevent this phenomenon.

b. The size of the sensor – size does matter. The bigger the sensor per given resolution, the larger surface area each photoelectric cell (pixel) is given, which eventually means each cell can gather more photons with less interference from other cells. This mainly means less "noise" in the image. A smaller sensor also requires a higher lens resolution since the pixels are smaller and denser. Another thing affected by sensor size is the Depth of Field, a smaller sensor means Greater DOF and larger sensor is shallower DOF. Smaller sensor also has a "crop factor multiplier" because it has a smaller surface than a 35mm camera film, it takes a small portion of the projected image (crops the image) and produces a normal frame, hence practically is multiplying the focal length covrage, for example Nikon DSLR has a X1.5 crop factor and Canon DSLRs have X1.6 crop factor And 50mm lens on a DSLR will produce similar frame coverage as 75mm lens on film.

c. CMOS – Complementary Metal Oxide Semiconductor – now that's a long name… in short, CMOS were originally used in computer chips, it's much cheaper to produce and the main difference between CMOS and CCD is that in CMOS each photodiode (pixel) holds several transistors that read and amplify the signal. Because light also hits these transistors and not only the photoelectric cells, these sensor are traditionally less sensitive to light. (I suggest describing CMOS before CCD)

d. CCD – Charge-Coupled Device – one of the two common types of sensors. In CCD sensors, the photoelectric cells on the sensor don’t have a designated transistor, but rather deliver the electrons flow to the side of the sensor, where circuitry surrounding the sensor processes the output and sends it to an A/D converter. CCD sensors produce high image quality, yet are more expensive to manufacture and consume more energy than CMOS. CCDs are more common than CMOS sensors.

e. ISO – International Standard Organization a.k.a. ASA American Standard Association. The sensor's sensitivity is actually the amount of amplification of the signal from the sensor. Every sensor has a base sensitivity in which it provides the 'cleanest' photo, and each increase in that sensitivity will shorten the exposure and produce a noise side affect. In low light conditions it's recommended to use higher ISO, but in general lower ISO is recommended for avoiding this "noise"

f. Electronic Shutter – we've already talked about regular shutters, now we'll talk about the electronic one. Some cameras only posses a mechanic shutter, some only an electronic shutter and some have both. Electronic shutter defines the length of exposure by stopping the readout from the photoelectric cells after the specified period of time.

g. Noise – just as there are interferences when we listen to the radio, there are interferences in digital photography. In the sensor, as we amplify slightly the sensor's signal in low ISO, we will find an image clean of noise. But as we raise the ISO, more aggressive amplification is occurring and static interference between the photoelectric cells is also amplified, producing random pixels in different colors. It's recommended to shoot bright images in high ISO in order to reduce noise visibility. Another kind of noise is produced in long exposure shots, which is due to sensor heating. Try avoiding long exposures that produce this noise by taking several shorter exposures and stitching them together via software.

h. Noise Reduction – every camera applies some sort of noise reduction. On the computer you can use software such as Noise, Ninja, or Neat Image to clean this noise. In long exposure shots, you may apply the camera’s noise reduction method, entitled "dark image subtraction" - in which the camera takes another frame after the original frame, this time with a closed shutter (resulting in a dark frame), and then reduces the "hot pixel" in the dark frame from the original one.

i. A/D Converter – we mentioned that the sensor receives photons and emits electrons in reaction, but the processor in the camera doesn't know how to read this information and that's why we have the A/D converter which reads the voltage output for each pixel from the sensor and attaches a value between 0 and 255 (in case of 8 bitcolor depth) and then gives it a binary sequence so that the CPU can process it. In 8 bit depth, 0 is black and 255 is totally white. In 12 bit system the pixel is given a value between 0 and 4096 (2^12) , resulting in a greater dynamic range and more details in mid-levels.

j. Color – as we said - each cell can distinguish 256 levels of brightness, so how does color get into the picture? By using a Color Filter Array of RGBG (Red-Green-Blue-Green) we have a quarter of the pixels red, a quarter blue and half of them are green. The reason for that is because the human eye has different sensitivity for each color. After the readout, an interpolation called Bayer's interpolation, is being done in the processor, which cleverly calculates the colors for each pixel in consideration with its surrounding pixels, This creates a 24-bit color-depth image (2^8 x 2^8 x 2^8 = 2^24). A different kind of color rendering is the Foveon way, which uses three different layers of color for each pixel in the sensor thus creating the most reliable color.

k. Buffer – after the image is created, it is stored in a temporary memory section called "buffer" until it has finished its processing and is written to the memory card. The size and efficiency of the buffer affects the speed of photography and how many photos can be taken before we need to wait for the camera to take the following shot.

l. The Processor – one the most important parts of the digital camera. The manufacturers invest many resources in developing the main image processor and the software that creates the image and determines the final image quality. These processors are very powerful and can process tens of millions of pixels a second.

m. Memory Cards - at the end of the photography process, the image is saved on a flash memory card. It's very durable and very small and can hold thousands of photos (on an 8 GB card for example). The speed of the card determines how fast the buffer will empty itself and shortens the delay between each photo.

I Hope you enjoyed part II of Photography 101.

Join me in the next episode – The Lens System

Previous Parts – The Camera, The Digital System

Yours truly,

Roie Galitz

For more photography articles – composition, Depth of field, Filters, Flash Photography, infrared photography, Sunset photography, Blue Channel, Panning