Histogram Equalization algorithm is an image correction method that evenly distributes pixel's color intensity across the image. It makes too light regions darker, too dark regions lighter and adjusts the contrast. It can help to normalize low-contrast images and also images, which have both too dark and too light regions which make it impossible to correct them using either Brightness or Contrast filter. In particular, this can be important to adjust the image this way before the optical text recognition procedure, to make the text more readable. The next video shows how the "Equalize" filter is implemented in the Image Reader online text recognition service and how it can be used for an image, on which other filters could not help:

The definition of Histogram Equalizaton: https://en.wikipedia.org/wiki/Histogram_equalization