Floating-Point Arithmetic

We demonstrate a set of simple modifications to make a basic CCD implementation failure-proof. Using error analysis, we prove the safety of this method and we formulate tolerance values to reduce false positives.