English (English) - Change language

How hypotheses for Labeled Field elements are generated and assessed

A Labeled Field element is a compound element with pre-defined components. Hypothesis for this element are generated and assessed in the same way as for Group element.

A hypothesis generated for a Labeled Field element has the following properties:

Property Description
Element name The full name of the element.
Page The number of the page on which the element has been detected.
Surrounding rect The coordinates of the surrounding rectangle that encloses the region of the hypothesis.
Width The width of the region of the hypothesis.
Height The height of the region of the hypothesis.
Detected Indicates whether the area of the element has been detected (True) or a null hypothesis has been generated (False).
From the best path Indicates whether the hypothesis belongs to the best path in the tree of hypotheses (True) or not (False).
Pre-search quality Indicates the quality of the hypothesis required for the hypothesis to satisfy the properties of the element, specified either directly or via Advanced pre-search relations.
Post-search quality Indicates the quality of the hypothesis required for the hypothesis to satisfy the conditions in Advanced post-search relations.
Chain quality Indicates the current quality of the segment of the chain of hypotheses, from the first subelement of the current group to and including the current subelement. The quality of the chain is calculated by multiplying the qualities of all the subelements in the chain.Chain quality makes it possible to compare rival chains.

The hypotheses generated for subelements have all of the above properties supplemented with the following properties.

LabeledField.Name subelement

Property Description
Keyword Indicates the keywords included in the hypothesis. For each keyword, the number of errors is indicated.

LabeledField.Gap subelement

Property Description
Orientation Indicates the orientation of the detected gap.
Histogram maximum in search area Indicates the histogram maximum in the search area.
White Gap threshold Indicates the calculated threshold to treat the gap as detected.
Histogram maximum within hypothesis Indicates the histogram maximum in the region of the gap hypothesis.

LabeledField.Field subelement

A hypothesis generated for this subelement may have different properties depending on the type of field:

Field type Property Description
Any text Text The text of the hypothesis.
Fixed variants Keyword The keywords included in the hypothesis. For each keyword, the number of errors is indicated.
Number Text The text of the hypothesis.
Currency Value The detected numerical value.
Currency name The detected name of the currency.
Date Day Day.
Month Month.
Year Year.
Phone Number The detected telephone number.
Prefix The detected telephone prefix.
Regular expression Text The text of the hypothesis.

More:

Search area

Additional search constraints

10/9/2020 8:50:41 AM


Please leave your feedback about this article