A modern phone camera is mostly software wearing a lens. The hardware is impressively small; the magic happens in processing.
The hardware path
Light enters through the lens and lands on an image sensor — millions of tiny light-collecting wells called photosites. Each one measures how much light hit it.
- A colour filter over the sensor lets each photosite see only red, green or blue.
- The chip reconstructs full colour by interpolating between neighbours.
Why bigger pixels matter
Larger photosites collect more light, which means less noise in dim conditions. That is why a phone with fewer megapixels can sometimes beat one with more — pixel size often matters more than pixel count.
Computational photography
This is where phones pull ahead:
- The camera captures several frames the instant you tap the shutter.
- Software aligns and merges them to reduce noise and widen dynamic range.
- Machine learning sharpens detail, balances colour and brightens shadows.
The photo you see was never captured in a single instant — it is the best blend of many.
The trade-offs
All that processing can look over-sharpened or unnaturally bright. Knowing it is happening lets you dial it back and capture a more honest image.
