Browse Prior Art Database

Cognitive evaluation of elements worn out Disclosure Number: IPCOM000250159D
Publication Date: 2017-Jun-07
Document File: 3 page(s) / 323K

Publishing Venue

The Prior Art Database

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 54% of the total text.

TITLE: Cognitive evaluation of elements worn out


Every device and its elements have their time of life. Elements and whole devices/machines

when being used, wear out, but we can extend time of life of machines, when replacing only

worn out elements. Problem is that we know that we have to replace that element only when it

fails or starts working incorrectly.

We are proposing usage of cognitive system for analysis of pictures of elements of

device/machine to suggest when and which elements should be fixed/replaced to avoid accidents

of whole device/machine.

We are proposing of usage of cognitive system such as machine learning, combined with image

recognition and processing to recognize which elements may fail in future.

Our idea is based on two steps – learning of cognitive system and usage of it.

There should be created knowledge base, created from pictures of newly manufactured

devices/machines. Then those pictures should be processed through some pattern recognition, for

example Edge detection algorithm, to identify elements. Then employee of manufacturing

company should only name detected parts, so our cognitive system will be able to match that

detected element to its name.

We are introducing idea of usage of surveillance camera inside devices. There many solutions,

that are using cameras to observe devices (i.e. server) but only from outside. First image from

new device/machine would be picture that will be basic picture for learning the cognitive system,

because we can assume that new device has all parts that are in good shape and are working

properly. There should be minimum one camera inside, but better solution would be having

multiple micro cameras inside, that will be able to observe every element inside the device.

Learning process of cognitive solution will begin with initial pictures of newly manufactured

device, then cognitive solution should recognize...