Mozilla Introduces The mozjpeg Encoder
Mozilla is trying to create an efficient JPEG encoder with a a new project called 'mozjpeg'.
JPEG has been in use since around 1992 and it remains the most popular lossy compressed image format on the Web. However,
The number of photos displayed by the average Web site has grown over the years, as so has the size of those photos. Reducing the size of these files in order to save bandwidth is an obvious goal for optimization.
So Mozilla started the 'mozjpeg' project. Available today as version 1.0, it is a fork of libjpeg-turbo with 'jpgcrush' functionality added. Many people have been reducing JPEG file sizes using a perl script written by Loren Merritt called 'jpgcrush'. It losslessly reduces file sizes, typically by 2-6% for PNGs encoded to JPEG by IJG libjpeg, and 10% on average for a sample of 1500 JPEG files from Wikimedia. It does this by figuring out which progressive coding configuration uses the fewest bits. Since no production encoder has this functionality built in, Mozilla added it as the first feature in 'mozjpeg'.
Mozilla's next goal is to improve encoding by making use of trellis quantization.
The number of photos displayed by the average Web site has grown over the years, as so has the size of those photos. Reducing the size of these files in order to save bandwidth is an obvious goal for optimization.
So Mozilla started the 'mozjpeg' project. Available today as version 1.0, it is a fork of libjpeg-turbo with 'jpgcrush' functionality added. Many people have been reducing JPEG file sizes using a perl script written by Loren Merritt called 'jpgcrush'. It losslessly reduces file sizes, typically by 2-6% for PNGs encoded to JPEG by IJG libjpeg, and 10% on average for a sample of 1500 JPEG files from Wikimedia. It does this by figuring out which progressive coding configuration uses the fewest bits. Since no production encoder has this functionality built in, Mozilla added it as the first feature in 'mozjpeg'.
Mozilla's next goal is to improve encoding by making use of trellis quantization.
