Home · All Namespaces · All Classes · Grouped Classes · Modules · Functions

User Interface: Input Methods: Stroke Recognition: Full Screen Stroke Recognition
Functional Description

The Full-screen Handwriting Input plug-in allows the user to draw characters for text input over the top of dialogs. In addition the plug-in also allows the user to customize it to their requirements.

Features

• Dictionary Selection
• Input Hinting
• Input Mode
• Next Word
• Smart Activation
• Symbol Input
• Word Recognition
• Word Selection

Dictionary Selection

The user is able to specify an alternate language to be used for
word recognition.

Input Hinting

Full-screen Handwriting is able to take input hints from text input
elements in applications and use the input hint information to improve
accuracy in recognizing characters for that text input.

Input Mode

The user is able to specify the type of input apart from that
suggested by the text element, and hence modify how the plug-in recognizes
drawn characters.

Next Word

The user is able change the recognized word to the next most likely
word.

Smart Activation

Full-screen Handwriting is only active when an element that accepts
text input has focus. When it is active it:

• Passes screen taps through to the underlying application unless they are made immediately following a previous stroke.
• Passes screen strokes through to the underlying application if they are preceded by a pausing movement of the stylus for a short period of time.

Full-screen Handwriting always ignores strokes that are started
within the area designated for a scrollbar on the right side of the screen.

Symbol Input

The user is able to open a pop-up of additional symbols.

Word Recognition

Full-screen Handwriting prioritizes the recognition of characters as part
of its attempts to find a matching word. If a drawn character closely matches two or more characters,
then the plug-in biases towards characters that make up complete words.
For this reason a set of digits is also considered to be a word.

If no word is matched, then the plug-in uses the most closely matched
characters as they are drawn.

For example, a character drawn on the screen might be either an 'n' or an 'h'
and even though it might most closely resemble a 'h' the word recognition
feature waits to determine the character until the second character
is entered. If the second character is an 'o' then the application
assumes that the first character was an 'n'. However if a third character
is added such as 'w' then since the first character more closely resembled
a 'h' then the recognition feature suggests 'how' and provide 'now' as
a secondary option.

If characters are added until no word in the dictionary matches, then the
application reverts to spelling the word with 'h' since it was the most likely letter.

Words consisting of more than two characters that are not in the dictionary are added to the list of recognized words.

Word Selection

The user is able to open a pop-up of all possible matches to
the set of characters they have drawn.