LPR matcher technique: Common and contiguous characters - Security Center 5.8

Security Center Administrator Guide 5.8

Security Center 5.8

The LPR matcher uses the "Common and contiguous characters" technique to improve plate read accuracy rate (sometimes called “fuzzy matching”).

NOTE: This method is used for overtime parking enforcement only.

You can configure how the LPR matcher handles common and contiguous characters by modifying the MatcherSettings.xml file. For more information, see MatcherSettings.xml file.

NOTE: You can also configure the following LPR matcher techniques in the MatcherSettings.xml file:

The following settings are available when configuring common and contiguous characters:

Necessary common characters
The minimum number of characters that need to be common to both the first and second plate read. The characters must also appear in the same order in the plate, but not necessarily in sequence.
Necessary contiguous characters
Minimum character sequence length between the first and second plate read.

In overtime enforcement, there is an extra margin of error because the LPR matcher is comparing a plate read against another plate read, not against a hotlist or permit list created by a person.

Here’s an example with the LPR matcher configured to look for five common characters and four contiguous characters (default). The LPR matcher also allows for the default one OCR equivalent character, which can count as a common or contiguous character.

Plate read 5ABC113 matches with 5A8CH3 (example 1) and 5ABCH3 (example 2) because the following conditions are met:
OCR equivalence
The OCR equivalents B and 8 are considered the same character and apply towards the common and contiguous character count.
Five common characters
Both reads have 5, A, B/8, C, and 3 in common, and they all appear in the same order. The “3” is not in sequence, but it respects the order.
Four contiguous characters
Both reads have 5, A, B/8, and C in sequence.

Plate read 5ABC113 does not match with SA8CH3 (example 3) because there are two OCR equivalents in the second read (S/5 and B/8). You allowed for only one OCR equivalent.

Using common and contiguous characters helps reduce the margin of error involved when both first and second plate reads are coming from the Sharp.