Google Increases JPEG Compression by 35% Without Loss in Image Quality

The company's Guetzli JPEG encoding algorithm improvement effort looks to have a greater immediate impact because of its ability to potentially reduce image file sizes by up to 35% at the same image quality level when compared to traditional encoders, while maintaining decoding compatibility with existing browsers. However, while Guetzli creates smaller image file sizes, the tradeoff is that these search algorithms take significantly longer to create compressed images than now available methods. This will enable webmasters to develop webpages that can load faster while using comparatively less data.

The new open source algorithm is called Guetzli, which is Swiss German for "cookie".

Uncompressed original image (left), libjpeg-encoded JPEG (middle), Guetzli-encoded JPEG (right). The advantage to using Guetzli instead of a new format is that the images are still regular JPEG files, and so they're still compatible with nearly every browser and application that exists.

More news: Somali refugees die in Yemen attack

The encoding process for a JPEG can be broken down into six distinct parts: color space transformation, downsampling, block splitting, discrete cosine transform, quantization, and entropy coding. That may seem like a deal-breaker but because of how well it performs in the quality department, Google feels the slower compression time is a worthy trade-off. It uses an advanced psychovisual model that attempts to strike a balance between small file sizes and image fidelity by honing in on the kind of details that the human eye is drawn to, according to a post published to the Google Research Blog. The algorithm used by the company allows it to compress JPEGs by up to 35% more than now available methods. It would be interesting to see if Guetzli scores a wider acceptance.

Google has made the Guetzli encoder open source, and it's available on Github for anyone to integrate into their own projects or to use on their own. For example, a multi-megabyte image stored in BMP or PNG format can look nearly exactly the same converted to a JPEG that's only a few hundred kilobytes in size.

Furthermore, Google claims that Guetzli does this compression without compromising on image quality. "Although Guetzli may be too slow for many practical uses, we hope that it can show direction for future image format design", the researchers said.