Browse Prior Art Database

Image Recognition Based Installation

IP.com Disclosure Number: IPCOM000198701D
Publication Date: 2010-Aug-12
Document File: 3 page(s) / 41K

Publishing Venue

The IP.com Prior Art Database

Abstract

Most software installation programs have two options: manual installation and silent installation. For silent installation, a user has to predefine a set of configurations and once the silent installation starts, the user has no control over the process. For manual installation, because the software systems keep getting more and more complex, sometimes it takes a long time to install; sometimes the user steps away during the long installation process, while the installation is waiting for an input for some configurations. Also, sometimes users may overlook and don't catch some installation errors such as display errors. This invention will be extremely helpful to software developers, testers, L2 and L3 support engineers, and customers. For software developers and testers who often have to install software, this patent idea will significantly reduce the amount of time spent interacting with software that takes hours to install. For L2 and L3 engineers, this patent idea can be used to capture any installation mistake or difference that customers have. For example, our product ITM has lots of installation issues in the field due to the complexity of the product and product install. For customers, this patent idea also can be used for any new version software installation to help the users to understand any new feature of the latest version of the software. It also can help with the customers on non-production environment setup. The software GUI installation normally has a set of images, our patent is images recognition based decision making for actions during software installation, which can be applied to various software installation, unlike a scripted automation, our patent is semi-automated software install, users have better control of the automated GUI based installation, this disclosure proposed install can be configured to do partial automation, partial manual install, it can also detect GUI display problem which this prior art and other scripts based install automation can't do. The install log has image captures, it captures image difference or new image or GUI display error, user manual input then save into the installation log. Our patent gives users better control over the automation, it recognizes the image and can be configured to wait for user's inputs in order to continue the installation, or configured to simulate keystrokes, mouse movement to automatically continue the installation.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 53% of the total text.

Page 1 of 3

Image Recognition Based Installation

This invention is to automate software installation based on (graphical) image recognition. The software GUI installation normally has a set of images,
users can reuse or predefine a set of images to be recognized such as the older version of the software installation images,
users then define a set of rules such as

when image A is recognized, this automated GUI installation stops and waits for user's input,
when image B is recognized, the installation will trigger menu button click such as "Next", when image C is recognized such as the version of the software or existing versions of the software in a same or different location, the installation captures the new image and stores it to a log file.
when image D is recognized such as a runtime error, abort the installation.

When image E is recognized for remote deploy, it can pull the result from the remote deploy and save it to a file for users to verify.

When the automated GUI installation starts, it recognizes the images and compares the differences, based on the users defined rules, it either continues the installation by triggering the menu button click or waits for the user input to manually operate certain selections such as what component to installation. The automated GUI installation finally output the new images, different images, user manual input results into a log file for post-installation review, therefore, users can see what happens during the installation such as whether...