Interference No. 103,587
Levien’s Case for Priority
The Evidence:
The inventor, Raphael Levien, testified to the following effect.
In 1987, Raphael Levien lived at home with his parents. At that residence, he had lab
equipment which consisted of a digital camera to capture digital images, a 286 computer to
screen an image from the camera, and a laser printer for printout of the screened image. This
equipment allowed him to make a printout of a captured image on the printer.
Dot 1 was a technique developed by Levien for printing out images to create halftones by
computing an initial random halftone, then iteratively refining it by computing an error image,
filtering the error image with a simple FIR filter, then adding the filtered error image to threshold
values for computing the next iteration of the halftone. Eventually, after about 16 hours of
iterations, the process would converge. By adjusting the parameters of the filter, it was possible
to create a wide variety of effects, including variable size dots. However, graininess from the
initial random halftone was a problem. Dot 1 also took an unacceptably long time to screen an
image, that is, to process input signals from a digital camera to reproduce an image.
Nevertheless, Dot 1 was an improvement in error diffusion screening and despite its problems,
Dot 1 was incorporated into a complete commercial working system, which was actively
marketed.
On November 8, 1987, Levien told his father, Jack Levien, that he could form halftone
dots using prior outputs to create hysteresis in a recursive process and control the coarseness of
the halftone dots. He told his father that hysteresis would increase the size of groups of dots by
increasing the tendency of smaller dots to form a larger variable size dot. Raphael showed his
14
Page: Previous 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Next
Last modified: November 3, 2007