28-12-2012, 07:02 PM
CCD & CMOS Image Sensors
[attachment=45706]
What is an Image Sensor?
An Image Sensor is a photosensitive device that converts light signals into digital signals (colours/RGB data).
Typically, the two main types in common use are CCD and CMOS sensors and are mainly used in digital cameras and other imaging devices.
CCD stands for Charged-Coupled Device and CMOS stands for Complementary Metal–Oxide–Semiconductor
Uses?
Image sensors are not only limited to digital cameras.
Image sensors are used in other fields such as:
Astronomy, most notably in the Hubble Space Telescope
Machine vision/sensing
UV Spectroscopy
Etc.
How Image Sensors Work
Both CCD and CMOS sensors work by employing photosensitive circuitry that reacts to light and stores the analog signals as digital data, namely an image.
They both use different methods to achieve this.
First we will take a look at CCD image sensors.
CCD
A CCD, or a Charged-Coupled Device, is a photosensitive analog device that records light as a small electrical charge in each of its pixels or cells. In essence a CCD is an collection of CCD cells.
The signal captured by the CCD requires additional circuitry to convert the analog light data into a readable digital signal.
This is mainly layers of capacitors called Stages which act as a way to transport the analog signal to an array of flip-flops which store the data all controlled by a clock signal.
This is the definition of an Analog Shift Register.
When light strikes a CCD, it acquires an electrical charge according to how much light has hit the particular CCD cell.
Each CCD cell can transfer its charge to its neighboring cell and then off to external circuitry.
The charge is then read off by an analog-to-digital converter as an integer on a range of 0 to 4095 for most modern DSLR cameras. Lower ranges exist, such as 0-255, for lower quality cameras.
How CCDs Record Colour
Each CCD cell in the CCD array produces a single value independent of colour.
To make colour images, CCD cells are organized in groups of four cells (making one pixel) and a Bayer Filter is placed on top of the group to allow only red light to hit one of the four cells, blue light to hit another and green light to hit the remaining two.
The reasoning behind the two green cells is because the human eye is more sensitive to green light and it is more convenient to use a 4 pixel filter than a 3 pixel filter (harder to implement) and can be compensated after a image capture with something called white balance.
Ex. A Bayer filter applied to the underlying CCD pixel
CMOS
A CMOS, or Complementary Metal Oxide Semiconductor, each pixel has neighboring transistors which locally perform the analog to digital conversion.
This difference in readout has many implications in the overall organization and capability of the camera.
Each one of these pixel sensors are called an Active Pixel Sensor (APS).
[attachment=45706]
What is an Image Sensor?
An Image Sensor is a photosensitive device that converts light signals into digital signals (colours/RGB data).
Typically, the two main types in common use are CCD and CMOS sensors and are mainly used in digital cameras and other imaging devices.
CCD stands for Charged-Coupled Device and CMOS stands for Complementary Metal–Oxide–Semiconductor
Uses?
Image sensors are not only limited to digital cameras.
Image sensors are used in other fields such as:
Astronomy, most notably in the Hubble Space Telescope
Machine vision/sensing
UV Spectroscopy
Etc.
How Image Sensors Work
Both CCD and CMOS sensors work by employing photosensitive circuitry that reacts to light and stores the analog signals as digital data, namely an image.
They both use different methods to achieve this.
First we will take a look at CCD image sensors.
CCD
A CCD, or a Charged-Coupled Device, is a photosensitive analog device that records light as a small electrical charge in each of its pixels or cells. In essence a CCD is an collection of CCD cells.
The signal captured by the CCD requires additional circuitry to convert the analog light data into a readable digital signal.
This is mainly layers of capacitors called Stages which act as a way to transport the analog signal to an array of flip-flops which store the data all controlled by a clock signal.
This is the definition of an Analog Shift Register.
When light strikes a CCD, it acquires an electrical charge according to how much light has hit the particular CCD cell.
Each CCD cell can transfer its charge to its neighboring cell and then off to external circuitry.
The charge is then read off by an analog-to-digital converter as an integer on a range of 0 to 4095 for most modern DSLR cameras. Lower ranges exist, such as 0-255, for lower quality cameras.
How CCDs Record Colour
Each CCD cell in the CCD array produces a single value independent of colour.
To make colour images, CCD cells are organized in groups of four cells (making one pixel) and a Bayer Filter is placed on top of the group to allow only red light to hit one of the four cells, blue light to hit another and green light to hit the remaining two.
The reasoning behind the two green cells is because the human eye is more sensitive to green light and it is more convenient to use a 4 pixel filter than a 3 pixel filter (harder to implement) and can be compensated after a image capture with something called white balance.
Ex. A Bayer filter applied to the underlying CCD pixel
CMOS
A CMOS, or Complementary Metal Oxide Semiconductor, each pixel has neighboring transistors which locally perform the analog to digital conversion.
This difference in readout has many implications in the overall organization and capability of the camera.
Each one of these pixel sensors are called an Active Pixel Sensor (APS).