Pixel binning refers to the process of combining the readout values of adjacent pixels across an image after image capture. Binning works by adding together the electrical charges collected by neighboring pixels of the same color and reading them out as a single simulated pixel. Under low-light conditions, this improves camera sensitivity and image brightness. However, the downside is equally obvious: once pixels are merged, image resolution is significantly reduced, and this trade-off is largely unavoidable.
The "Detail Loss" Trade-Off
Today's mainstream pixel binning technologies include 4-in-1, 9-in-1, and even 16-in-1 pixel merging. Among them, 4-in-1 pixel binning is the most widely adopted solution.
In essence, pixel binning still relies heavily on computational processing. It combines four or more neighboring pixels into a larger "virtual pixel" through algorithmic reconstruction. Theoretically, this larger pixel behaves like a normal-sized pixel with standard spacing, improving light sensitivity and reducing image noise.
Using the relatively straightforward 4-in-1 solution as an example, this technology is also known as Quad Bayer.
Quad Bayer Technology
Within a single exposure cycle, Quad Bayer technology controls the exposure duration of each photosensitive unit in the same region, enabling two bracketed exposures in an L+L+S+S configuration. In practice, this means HDR-like functionality can already be achieved at the CMOS sensor level.
Sony's Quad Bayer color filter array first appeared commercially in the Huawei P20 Pro, released on March 27, 2018.
Compared with the traditional Bayer array, the structure of the Quad Bayer array is quite different. When four pixels are combined into one larger "virtual pixel," these larger pixels effectively recreate the pattern of a conventional Bayer array. After pixel merging, image resolution is reduced by approximately 75%. For example, a 48MP sensor effectively outputs a 12MP image after 4-in-1 binning.
Another way to understand Quad Bayer is that it essentially subdivides each traditional Bayer pixel into four smaller units, allowing for more refined light capture and processing.
Traditional Bayer vs Quad Bayer
A conventional Bayer array consists of:
- 50% Green pixels
- 25% Red pixels
- 25% Blue pixels
When Quad Bayer operates without pixel binning under normal lighting conditions, advanced reconstruction algorithms can still achieve full-color sampling equivalent to a traditional Bayer array within the same sensor area:
- 50% Green
- 25% Red
- 25% Blue
Why Pixel Binning Became Popular in Smartphones
Pixel binning is primarily a smartphone imaging technology. Features such as direct image output, HDR processing, and computational photography are major advantages smartphones have over traditional cameras.
Under suitable lighting conditions, Quad Bayer modes can indeed improve effective resolution to some extent. However, compared with conventional Bayer arrays, the improvement in dynamic range is often more noticeable.
In reality, pixel binning moves traditional post-processing techniques such as exposure stacking into the image capture stage itself. In low-light environments, it sacrifices resolution in order to compensate for insufficient light intake.
Although it cannot simultaneously maximize both ultra-high resolution and ultra-high dynamic range, pixel binning represents a practical compromise made by smartphone sensor manufacturers within the fixed physical size constraints of CMOS sensors.
Pixel Binning vs Pixel Shift
Before learning about pixel binning, many people may have heard of another imaging technology called Pixel Shift.
Pixel Shift works by moving the image sensor with sub-pixel precision between exposures. This allows every pixel position to capture complete RGB color information, overcoming the limitations caused by the Bayer color filter arrangement.
Pixel binning and pixel shift are fundamentally different concepts.
Pixel Shift improves overall image detail and signal-to-noise ratio through oversampling, ensuring that each pixel contains full RGB information.
Pixel binning, on the other hand, merges multiple pixels together to improve light sensitivity and low-light performance, usually at the cost of image resolution.
Different Approaches in Mobile Imaging
Some manufacturers have chosen a different direction by keeping sensor resolutions between 10MP and 20MP instead of aggressively pursuing ultra-high megapixel counts.
For example, both Google and Apple relied on 12MP CMOS sensors for many years. Apple maintained a 12MP sensor from the iPhone 6s through the iPhone 13 series, while Google's computational photography and Pixel Shift-like technologies helped create some of the strongest single-camera smartphone systems in the industry, such as the Google Pixel 3.
Ironically, manufacturers that did not heavily market ultra-high megapixel counts or extremely large sensors were often the ones that pushed smartphone photography to new heights.
