Image optimization is an art as well as a science. It is an art because there is no definite answer for how best to compress an individual image and science as numerous well-developed techniques and algorithms can considerably reduce the science of an image. For people who are building an app or a website, they are always looking for something to reduce the size of the image. Now, Google has come up with its latest image compression algorithm which reduces the JPEG file size by 35%.
Guetzli, the new image compression algorithm, works with existing browsers and image processing tools. According to Google, Guetzli allows for small file sizes without compromising much on the quality of the image and targets the quantization stage of the compression process. This is the point where files are normally squished with algorithms but result in a poor quality image.
The new compression algorithm by Google is slower than the available options for this purpose. It approximates color perception and visual masking in a more detailed manner than the rest. The system uses an algorithm to achieve better results.
In the above-given examples, on the left is shown the uncompressed original image, in the middle, there is the result of libjpeg and on the right is the image results of Guetzli. The results clearly show how Google’s solution has less ringing items in the above examples.
This open-source JPEG encoder is available to be downloaded and implemented in the projects.
Via: The Next Web