PSYCH 221 – Applied Vision and Imaging Systems
Class Project
High Speed Document Sensing
Georgios Georgiadis
Lily Huang
Guillaume Leseur
Nicolas Meunier
Supervised by: Peter Catrysse
Introduction and
methods:
The goal of this project is to study the possibility of real time
misprint detections on high speed digital printers. With the development of
these new high speed digital printers, a single misprint may have important
consequences and lead to reprinting thousands of pages. This may be costly in
time as well as in money, and that’s why we are studying the possibility of automatically
detecting in real time these misprints to be able to react directly and solve
the problem.
So in this problem we have:
- A file that we want to print and.
- The actual result of the printer.
And we need:
-
A sensor to
have an image of the printed result.
-
A model of
the printer to know from the file which image to expect.
-
A model of
the sensor to know the theoretical output of the sensor to compare the actual
output with.
-
An
algorithm comparing these two results and deciding whether there is a misprint
or not.
We can summarize that in
this schema:
So far we have implemented the
printer model to convert a pdf file
to map of reflectance in MATLAB, as well as the model of the sensor that we set
up in ISET. We have also studied what is the theoretical precision of the
sensor in distinguishing patches of different cmyk values and its spatial
resolution.
