In its presentation, Google has called the Camera equipped by new smartphone line Pixel as the best of the mobile segment, by referring to benchmarks conducted by independent evaluation criteria. Certainly not to the resolution of the sensor (12.3 megapixels), but for the quality of the images. The real strength is represented by software .

To know more about is Marc Levoy, research under the Research Division of bigg. The issue can be complicated for those who do not chew photographic terms, but suffice it to say that in the jargon is called noise that disorder which usually generates shooting in low light conditions to ISO high and sensors not its quality. The trick used by Google in Pixel is simple: if there is little light, you quickly realize more images and then combine the information saved in order to obtain enough data for reproduce optimally the darker areas.

Photographing a shadow area generates a lot of noise, because few photons arrive on those pixels. Thanks to mathematics, however, if you take nine pictures, the noise is reduced by three times the square root of the number of photos made. In other words, take pictures improves the quality of the result.


In a sense, it is a similar approach to that which is used for high dynamic range images, where you take shots taken at different exposures to better balance light and shadow. The software developed by Google then deals to align perfectly the objects and subjects immortalized, an essential detail, since the photos combined are inevitably made in different moments, albeit with minimal delay.

Staying on the theme, the HDR + Pixel and Pixel XL, contrary to what happens to other smartphone, can be left always active, since it does not cause lag in saving images. Another step forward allowed by the software, as Levoy explains.

The moment you press the shutter button, actually does not save any pictures, because this has already happened. In fact they are saving many images. What happens when you press the button is that you establish an exact moment in which to refer to combine the images already captured.