Browse Prior Art Database

System and method for bidirectional page issue identification

IP.com Disclosure Number: IPCOM000249149D
Publication Date: 2017-Feb-08
Document File: 8 page(s) / 251K

Publishing Venue

The IP.com Prior Art Database

Abstract

Our solution proposed is a four-step automation mechanism to process Non-reversible icons on bidirectional environment. Non-reversible icons are photos or logos or text in images. Use color distribution to identify photos, logo crawling to identify logos, and then use OCR(Optical Character Recognition) to identify text in images. Finally, provide a quick feedback mechanism to flip back wrong icons and learn from users. It has the following characteristics: 1) Mechanism to identify photos; 2) Mechanism to identify logos 3) Mechanism to identify text-only images 4)Mechanism to correct a wrongly-flipped icon and learn from the error.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

1

System and method for bidirectional page issue identification

Arabic, Hebrew, Urdu, and Farsi (Persian) are languages written from right to left, we call them "bidirectional" languages. using the Arabic or Hebrew scripts. Numbers and English letters appearing in those languages are written from left to right, as well as segments of Latin (or Cyrillic or Greek) text which may be embedded in this text. The dual directionality aspects of such bidirectional (Bidi) text are posing challenges to the way this text is processed and presented in computer applications.

In daily work of globalization testing, in the context of bidirectionality, the following are key notions: base text direction segments or directional runs ordering scheme and associated reordering methods symmetric swapping

In daily work of globalization testing, if it is a continuous delivery project, the workload is all manual and very time consuming, the check points are many and could not be all automated, it will be more difficult if it is an on-premise system or a mobile application, instead of a web application.

Our invention is about A new method of bidirectional checking and issue identification.

Problem

For websites supporting bi-directional languages, such as Arabic, Hebrew, Urdu, and Farsi (Persian), special checking is required to ensure that page contents are correctly rendered, and automation is preferred. However, how do you know if an icon is reversible?

Solution & Claims

2

Non-reversible icons are photos or logos or text in images. Use color distribution to identify photos, logo crawling to identify logos, and then use OCR to  identify text in images. Finally, provide a quick feedback mechanism to flip back wrong icons and learn from users.

The solution proposed is a four-step automation mechanism. It has the following characteristics: Mechanism to identify photos1. Mechanism to identify logos2. Mechanism to identify text-only images3. Mechanism to correct a wrongly-flipped icon and learn from the error4.

•In addition to the usual processing of element alignment and icon reverse, we conduct the following analysis to images on page

Color distribution to identify photos (non reversible)1.

Logo database scan to identify logos (non reversible)2.

OCR to identify text in images (non reversible)3.

Similarity check on feedback database (non reversible)4.

•Then provide additional feedback mechanism

•Instant feedback mechanism for correcting icon direction

•Save feedback into database for similarity check

Color distribution  FIG-1:

3

logo database  Logo: scan and save major logos in database for picture comparison, refer to FIG-2

Save common used logos into local logo DB(Database)1. Scan the application html image element and also save into logo DB2. Picture Comparison of #1 and #2 using image processing method3. Execute "Not Reverse" if any matches in #3 comparison result4.

FIG-2:

4

OCR for text in image  FIG-3:

5

Similarity check on fe...