Ars Technica: American painter Jackson Pollock is well known for his revolutionary “drip” painting technique. Although to the untrained eye his works may look like random explosions of paint on canvas, a computer algorithm shows that his paintings have a distinctive and identifiable style. Lior Shamir of Lawrence Technological University has scanned various Pollock paintings and extracted from each 4024 “numerical image content descriptors,” including fractals, Zernike polynomials, Haralick textures, and Chebyshev statistics. A computer algorithm uses those descriptors to detect details and patterns unique to the artist in question. According to Shamir, whose study has been published in the International Journal of Arts and Technology, the computer program was able to differentiate between an original Pollock and a fake 93% of the time. Shamir’s source code is publicly available, and he says the software can be used to identify works of other artists.