Image Sensors – CMOS or CCD?

In a previous post we spoke about HD, IP, IR, POE, PTZ, theABC’s of IP cameras. But there are also C’s, as in CCD and CMOS. 

 CCD stands for Charge Coupled Device, and CMOS stands for Complimentary Metal Oxide Semiconductor. Both are image sensors that are used in IP cameras.  In pre-digital days a lens gathered light and focused it on film. In today’s world image sensors have taken the place of film. 

An image sensor is a silicon chip on whose surface is embedded millions of photosensitive diodes called photosites. Photosites collect light in the form of photons. The more light that hits a photosite, the more photons it records, thus photosites that hold light from highlights in the scene will have many photons while those  recording light from shadows will have few. 

 Each photosite will eventually contribute a single pixel to the image but first the sensor has to convert the photons into an electrical signal. Both CCD and CMOS perform this process but each does it in a different manner

·         CCD:  The first row of pixels is transported to a corner of the chip called a read-out register, then on to an amplifier, and finally to a digital converter where each pixel receives a digital value. Upon completion, the charge of the row is deleted and the row above it drops down to undergo the same procedure, to be replaced by the next and the next until all the pixels have been converted. When we say charge coupled device, charge coupled refers to the manner in which each row is coupled to the one above it.

·       

              CMOS    Each pixel in a CMOS chip is surrounded by several transistors that amplify and move its charge using traditional wires therefore allowing each pixel to be read individually. CMOS stands for complementary metal oxide semiconductor and refers to the process by which the chip is manufactured, with the circuits and diodes etched into its surface.This is the same process used to make computer processors and memory chips.

While CCD sensors and CMOS sensors perform the same function, the different processes by which these sensors are manufactured result in a few differences affecting both the cost and the image.  

Cost

CCD sensors are more expensive because the manufacturing process is specialized and expensive and can only be used to make CCD’s. 

CMOS sensors, on the other hand, are made using the common high yield process used for semi-conductors and memory chips all over the world.

 Mass manufacture on the same equipment cuts the cost of producing CMOS chips dramatically and spreads the fixed costs of the plant over a much larger number of devices. Costs are lowered even more because CMOS image sensors can have processing circuits created on the same chip whereas on CCD’s these processing circuits must be on separate chips.

Image

CCD sensors produce a higher quality image because they transport the charge across the chip without any distortion resulting in a minimum of digital noise.    

CMOS sensors tend to be more perceptible to noise plus the proximity of each pixel to several transistors results in photons hitting the transistors instead of the photosite, thus lowering light sensitivity. 

How big is a pixel?

It is not until the very end of digital conversion that a pixel can be said to have size or dimension. Each photosite on the image sensor has size and dimension, but the pixels themselves are just photons.

Even after they have been converted to electrical charges they lack size and shape. Think of the static charge you pick up when walking across a carpet on a dry w inter day. Can you see or measure it? And once it has been converted to digital code, it is but a series of zeros and ones. Think of your birth date or phone number. Can you give it a size or a shape?

No, a pixel must wait for the device that prints or displays the pixel before it takes on size and shape.  To do this a computer divides the screen or printed page into a grid of pixels. It then uses the values stored in the digital code to specify the brightness and color of each pixel in this grid. Think of painting by number.

 




 
How big is a sensor?

This is tricky as well. Think of standard resolution; let’s take XGA, which is defined by the width and height dimensions in pixels, or 1024x768. A sensor’s size is specified in this manner but since a captured pixel has no physical size, the sensor size is expressed in the number of photosites on its surface. However, this can be a virtual approximation of the number of pixels since in most cases each photosite will capture one pixel.

When it comes to determining what type of camera fits your needs it is not so much a question of CCD or CMOS but of resolution. And it is here that the size of the sensor and the total number of pixels comes into play. And that is a matter we will address in our next post.  

 But if you need to know now, call us at 800-431-1658 or use our contact form to request information.