mostly low frequency

Statistical analysis of digital images of paintings is a new method that that has recently been used to detect forgeries. Using a technique called wavelet decomposition, a picture is broken down into a collection of more basic images called sub-bands. These sub-bands are analyzed to determine textures, assigning a frequency to each sub-band. The broad strokes of a surface such as a blue sky would show up as mostly low frequency sub-bands whereas the fine strokes in blades of grass would produce high frequency sub-bands.[8] A group of thirteen drawings attributed to Pieter Brueghel the Elder was tested using the wavelet decomposition method. Five of the drawings were known to be imitations. The analysis was able to correctly identify the five forged paintings. The method was also used on the painting Virgin and Child with Saints, created in the studios of Pietro Perugino. Historians have long suspected that Perugino painted only a portion of the work. The wavelet decomposition method identified that at least four different artists had worked on the painting.