The technology behind the image and video optimization solutions was developed by Beamr (JPEGmini's parent company) 

JPEGmini's optimization is done by removing perceptually redundant information from an image file. This information is imperceptible by humans, and identified by the Beamr Quality Measure (BQM). 

BQM – a mathematical algorithm, which is highly correlated with the way humans perceive image quality. BQM can tell us with a very high level of accuracy, whether any visible artifacts exist in our optimized images when compared to the original images. These different attributes are analyzed for each of the image “tiles”, and combined into a single “quality score” for that tile. The tile quality score is then used to control the amount of compression applied to that tile.

Essentially, JPEGmini takes the original JPEG image and applies higher levels of compression to produce several candidate images that are more compressed. Then JPEGmini uses BQM to compare the quality of the candidate images. Finally, JPEGmini determines which is the smallest candidate (in bytes) that does not exhibit any additional visible artifacts when compared to the original JPEG. This is the selected candidate that JPEGmini outputs as the final optimized JPEG.  

Read more details through our Blog Article