Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Frame based handwriting case sensitive recognition method

IP.com Disclosure Number: IPCOM000216929D
Publication Date: 2012-Apr-25
Document File: 6 page(s) / 65K

Publishing Venue

The IP.com Prior Art Database

Abstract

This article introduces a frame based handwriting case sensitive recognition method in smart phones and tablet computers. This method is trying to address the problem of the automatically recognition of uppercase and lowercase characters for handwriting input method with a frame in the input area.Basically, a method is provided that will automatically recognize the uppercase and lowercase letters based on a restriction area for lowercase letters. When the instrument is powered on, user will see a frame inside (lowercase input area) the writing area (recognition area). And if user wants to input a lowercase letter, just write the letter inside the frame boundary. If user wants to input an uppercase letter, just put the letter bigger then the frame.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 6

Frame based handwriting case sensitive recognition method

This method is trying to address the problem of the automatically recognition of uppercase and lowercase characters for handwriting input method with a frame in the input area.

Nowadays, the smart phones and tablet computers become more and more popular. The handwriting input on those mobile devices is widely used by a large number of users. When writing English and some other languages, users might want to type uppercase or lowercase letters. And some of the letters have similar shape between the uppercase and lowercase except the size difference. E.g. C-c, O-o, S-s.

Let's take a real case to illustrate the problem. When we type 'c' which is the last character of the word"Basic", I want to have a lowercase 'c' rather than a uppercase. In current smart phones, they don't provide a very effective way to automatically recognize the uppercase 'C' and lowercase 'c'. Instead, the user needs to select from 2 candidates: one is the uppercase 'C', the other is the lowercase 'c'. But users need a smarter solution to avoid the selection process. That's what this method tries to address.

Besides a selection process, some existing handwriting recognition methods try to solve this problem in two ways. One is to compare input with a reference character set, and the comparison parameters can be adjusted to the input context or history dynamically. And the other is to compare the size of the input

with the sizes of other input characters in the input context that has no ambiguity in their cases. The drawbacks of such solutions are, firstly, the computing processes are complex which slows down the recognition. And secondly, the users are not aware of the underlying recognition method -- there is no way that they can adjust the parameters intentionally, and the recognition results often subject to errors with reasons unknown to the users. Both these drawbacks lower the usability of the handwriting recognition methods. Given our solution, the result is easy to compute, the users can customize the method themselves, and there is a more straightforward UI that would largely decrease chance of recognition errors.

Basically, we provide a method which...