Detail Man
New Member
If you happened to have read my post at:
viewtopic.php?f=53&t=3898&start=40#p19564
and/or have an interest in how CCD (Charge Coupled Device) image sensors transfer the individual photo-sensor data to a digital camera's hardware, this following information may interest you.
My post (link above) stated up-front that it was "mere speculation". There are a few things that I said in that post that were incorrect/incomplete. In the interest of providing valid (and non-misleading) information, I have made some significant edits of the related content of that post, in order to comport with the information presented below. I limit the "focus" to CCD (and not to CMOS) image-sensors (as the image-sensors in the cameras that most of us own are likely CCD-type). However, many of the same articles linked to below (also) present information about CMOS image-sensors, and how they operate, as well.
Starting with more brief and general material, and progressing to more in-depth descriptions:
"... to read the value (accumulated charge) of each cell in the image. In a CCD device, the charge is actually transported across the chip and read at one corner of the array. An analog-to-digital converter turns each pixel's value into a digital value."
http://electronics.howstuffworks.com/question362.htm
"CCDs are so named for the way they transfer charges between pixel wells, and ultimately out of the sensor. The charges are shifted from one horizontal row of pixels to the next horizontal row from top to bottom of the array. This is a parallel (or vertical) shift register architecture, with multiple vertical shift registers used to transport charges vertically down the rows. The charges are "coupled" to each other (thus the term charge-coupled device) so that as one row of charge is moved vertically, the next row of charge (which is coupled to it) shifts into the pixels thus vacated.
With the charges shifted down the parallel array row by row, you might wonder what happens to the charges in the last row of the sensor device. Using a serial shift register architecture, the last row is actually a horizontal shift register. Charges in that row serially transferred out of the sensor using the charge-coupling technique, making room for the next row to be shifted out, and the next, and so on. This serial transfer of charge out of the CCD is often described as a "bucket brigade," referring to its similarity to the old-fashioned fire department's bucket brigade.
Before being transferred out of the CCD serially, each pixel's charge is amplified resulting in an analog output signal of varying voltage. This signal is sent to a separate off-chip analog to digital converter (ADC) and the resultant digital data is converted into the bytes that comprise the raw representation of the image as captured by the sensor, prior to any post-processing. Unlike computer RAM that represents a 1 or 0 by either storing a charge or not, the charge on a CCD remains in analog form until the ADC stage late in the process."
http://www.extremetech.com/article2/0,2 ... 451,00.asp
Even more detailed information at:
http://en.wikipedia.org/wiki/Charge-coupled_device
Additional information sources:
(Somewhat simplistic, but perhaps helpful) Illustrations relating to CCD image-sensor charge-transfer architecture [CCD Principle (1) on Page 3, and CCD principle (2) on Page 4]:
http://www.harvestimaging.com/pubdocs/59a_vlsi-tsa.pdf
An instructive (although somewhat dated 2001 article about CCD vs. CMOS image-sensor technologies):
http://www.dalsa.com/public/corp/Photon ... willer.pdf
viewtopic.php?f=53&t=3898&start=40#p19564
and/or have an interest in how CCD (Charge Coupled Device) image sensors transfer the individual photo-sensor data to a digital camera's hardware, this following information may interest you.
My post (link above) stated up-front that it was "mere speculation". There are a few things that I said in that post that were incorrect/incomplete. In the interest of providing valid (and non-misleading) information, I have made some significant edits of the related content of that post, in order to comport with the information presented below. I limit the "focus" to CCD (and not to CMOS) image-sensors (as the image-sensors in the cameras that most of us own are likely CCD-type). However, many of the same articles linked to below (also) present information about CMOS image-sensors, and how they operate, as well.
Starting with more brief and general material, and progressing to more in-depth descriptions:
"... to read the value (accumulated charge) of each cell in the image. In a CCD device, the charge is actually transported across the chip and read at one corner of the array. An analog-to-digital converter turns each pixel's value into a digital value."
http://electronics.howstuffworks.com/question362.htm
"CCDs are so named for the way they transfer charges between pixel wells, and ultimately out of the sensor. The charges are shifted from one horizontal row of pixels to the next horizontal row from top to bottom of the array. This is a parallel (or vertical) shift register architecture, with multiple vertical shift registers used to transport charges vertically down the rows. The charges are "coupled" to each other (thus the term charge-coupled device) so that as one row of charge is moved vertically, the next row of charge (which is coupled to it) shifts into the pixels thus vacated.
With the charges shifted down the parallel array row by row, you might wonder what happens to the charges in the last row of the sensor device. Using a serial shift register architecture, the last row is actually a horizontal shift register. Charges in that row serially transferred out of the sensor using the charge-coupling technique, making room for the next row to be shifted out, and the next, and so on. This serial transfer of charge out of the CCD is often described as a "bucket brigade," referring to its similarity to the old-fashioned fire department's bucket brigade.
Before being transferred out of the CCD serially, each pixel's charge is amplified resulting in an analog output signal of varying voltage. This signal is sent to a separate off-chip analog to digital converter (ADC) and the resultant digital data is converted into the bytes that comprise the raw representation of the image as captured by the sensor, prior to any post-processing. Unlike computer RAM that represents a 1 or 0 by either storing a charge or not, the charge on a CCD remains in analog form until the ADC stage late in the process."
http://www.extremetech.com/article2/0,2 ... 451,00.asp
Even more detailed information at:
http://en.wikipedia.org/wiki/Charge-coupled_device
Additional information sources:
(Somewhat simplistic, but perhaps helpful) Illustrations relating to CCD image-sensor charge-transfer architecture [CCD Principle (1) on Page 3, and CCD principle (2) on Page 4]:
http://www.harvestimaging.com/pubdocs/59a_vlsi-tsa.pdf
An instructive (although somewhat dated 2001 article about CCD vs. CMOS image-sensor technologies):
http://www.dalsa.com/public/corp/Photon ... willer.pdf
