Can downsampling or pixel binning improve low-light image quality?
Asked 12/9/2012
2 views
2 answers
0
If lower final resolution is acceptable, can a low-light image be improved by combining each 2×2 block of pixels into one larger pixel, effectively making the output half the width and height? In theory, four small pixels should collect about the same total light as one larger pixel area. Does this reduce noise or make the image effectively “brighter,” and is it best done from RAW data rather than JPEG? Have cameras actually used this approach?
Originally by Photography Stack Exchange contributor. Source · Licensed CC BY-SA 4.0
Photography Stack Exchange contributor
13y ago
2 Answers
4
Yes there have been plenty of cameras that have aggregated pixels in order to reduce noise, from my ancient Fuji Finepix s602z, which would output one megapixel images above ISO400 to the very latest PhaseOne medium format backs which offer higher ISO settings in "sensor+" mode but at one-quarter of the number of pixels (half the width and height).
It's important to note that whilst theoretically you get as much light by summing the values of four small pixels, compared to one big pixel, in reality there is usually a small amount of read noise when getting the value so you get 4 times the read noise with smaller pixels. Some sensors perform the addition (known as binning) right on chip prior to readout and digitization in order to combat this.
Originally by user1375. Source · Licensed CC BY-SA 4.0
user1375
13y ago
0
Generated from our catalog & community — verify before relying on it.
Yes. This idea is real and is generally called pixel binning when the sensor combines neighboring pixels, or downsampling when it’s done later in processing.
Combining a 2×2 block does not create new light, but it can improve signal-to-noise ratio at the cost of resolution. Four pixels over the same total area collect about the same total photons as one larger pixel area, so averaging/summing them can make noise less obvious in the lower-resolution result.
The key difference is read noise: if four small pixels are read separately, you also incur read noise four times. That’s why on-sensor binning is better than combining pixels after readout. Some cameras have done this specifically for better high-ISO performance, including Fuji models and medium-format backs with reduced-resolution high-ISO modes.
Doing it from RAW is preferable because JPEG has already been demosaiced, tone-mapped, compressed, and color-processed. RAW preserves the sensor data, so combining pixels there is more technically sound.
Related designs have also been used for dynamic-range benefits, not just low-light noise reduction.
Recommended products
UniqueBot
AI13y ago
Your Answer
Related Questions
How does smartphone pixel binning compare to larger native pixels?
Can a Bayer RAW file be converted to a lower-resolution RGB image without demosaicing?
Will a 42MP full-frame camera cropped to Micro Four Thirds match a 20MP Micro Four Thirds camera for bird detail?
Why is an image file often smaller than pixels × bit depth suggests?
Does sensor pixel size affect depth of field or the circle of confusion?