|
|
 |
The Principles of Definiens Cognition Network Technology® The underlying principle of Definiens Cognition Network Technology is that it is context-based. The technology identifies objects rather than just examining individual pixels. It then makes inferences about those objects by looking at them in context.
To emulate the human mind's cognitive powers, Definiens made a radical depature from conventional approaches. Using patented segmentation and classification processes, we have developed a robust method of rendering knowledge in a semantic network. The technology examines pixels not in isolation, but in context. It builds up a picture iteratively, recognizing groups of pixels as objects. Just like the human mind, it uses the color, shape, texture and size of objects as well as their context and relationships to draw the same conclusions and inferences that an experienced analyst would draw.
Though somewhat simplified, the following example illustrates the basic principles of the Definiens Cognition Network Technology and how it extracts intelligence from images.
Illustration:
With the addition of a knife and fork, the circle is immediately classified as a plate. The object – the circle – hasn’t changed, but the human mind uses context and relationships with other objects to make intelligent inferences about the objects it perceives. And this is precisely how Definiens Cognition Network Technology works.
Cognition Networks
Definiens Cognition Network Technology is the only technology that classifies and analyzes imagery using all the semantic information required to interpret it correctly. Rather than examining individual pixels in isolation, it distils meaning from the objects’ connotations and mutual relations.
Of course, Definiens Cognition Network Technology recognizes pixel information. But it adds significant value to this information by creating a powerful cognition network in a series of iterative segmentation and classification steps. This dramatically enhances the value of intelligence and information extracted from the image. What’s more, the extracted information is fully quantified and qualified to meet the user’s specific demands.
Capture Domain Expertise
The power of Definiens Cognition Network Technology is its ability to quickly and easily transform user knowledge and domain expertise into automatic routines. Definiens applies advanced machine learning concepts to train the system and render this knowledge in a production environment, enabling it automatically to process millions of images.
Definiens Cognition Network Technology can accomplish even the most challenging image intelligence tasks throughout the enterprise. Its open and flexible client/server architecture accommodates any number of clients and servers required to make the most of Enterprise Image Intelligence and meet the given business requirements.
Learn more about the Definiens’ product suite |