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

An algorithm/solution to convert handwriting diagram to vector diagram

IP.com Disclosure Number: IPCOM000213495D
Publication Date: 2011-Dec-15

Publishing Venue

The IP.com Prior Art Database

Abstract

Handwriting diagram is very useful for discussing or design, but it's not very easy to backup or to do further editing, so this solution is to convert a handwriting diagram into a vector which is very easy to back up and also very easy to edit with existing application(such as MS Visio or Adobe illustrator).

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 18% of the total text.

Title

An algorithm/solution to convert handwriting diagram to vector diagram

Inventor/Author

Yanbo Chu

Jin Huang

Weiwei Zhao

Summary

Handwriting diagram is very useful for discussing or design, but it's not very easy to backup or to do further editing, so this solution is to convert a handwriting diagram into a vector which is very easy to back up and also very easy to edit with existing application(such as MS Visio or Adobe illustrator).

Background

Programmers always want to draw diagrams on the whiteboard during the design/discussion time, diagram is very easy to understand, and also easy to show our ideas, but the problem is sometimes, after we got the diagram on whiteboard we still need to draw it again with Visio or such applications, because we want to edit it and to backup or share it with someone else.

Prior Art/Solutions

·         One of them is OCR:

The difference between OCR and our solution in that:

1. Different usage: OCR is for character recognition, our solution focus on shapes and diagram recognition,

2. Different algorithm: OCR wants to compare the input image of character with a pre-defined image directly, and to identify the image by calculate the similarity between the 2 images; our solution doesn’t compare image, we will analyze the input image and got shapes or elements on image by pattern recognition, so our solution doesn’t need a pre-defined baseline to do compare operation, which will may our solution more light and smart;

3. Handwriting: OCR cannot deal with handwriting image, while our solution could process it well;

·         Another one is Hand writing input:

The difference between handwriting input recognition and our solution in that:

Different algorithm: handwriting input recognition processes input at drawing time, because handwriting input recognition want to indentify the image element by your input gesture and sequence, so handwriting input need more information and have more limitation to process static image, furthermore handwriting input doesn’t need to concern about noise; our solution doesn’t need to process input at drawing time, because our solution could indentify shapes or elements from static image, so it could be more flexible.

Description

1.      Summary:

The target of our solution is to convert a handwriting diagram on whiteboard into electronic vector diagram which could be edited by Adobe illustrator or Microsoft Visio easily, so there will be three main phases in our solution:

i)        Image Processing: Take screen-shot of the handwriting diagram, and then we  will do a series of image processing operations in order to get a skeleton drawing(picture-1) of handwriting diagram which is used for pattern recognition,

Picture-1 skeleton drawing of handwriting diagram

ii)      Pattern Recognition: The result of image processing should be a skeleton drawing (that mean the image is only one pixel width) , and then we will do a series of pattern recognition operations in order to get all lines on the image(picture-2), a...