The human eye recognises patterns better than computers. The computer crunches numbers better than people. Any quantification of images is therefore the basis for processing images in bulk.
I have been told that images cannot be quantified. My ‘number acrobatics’ do it.
What does it take to convince you? Are the samples in the header banner sufficient? Possibly not. For your mind will search fro meaning when it looks at pictures.
At the same time, your mind may not be capable to imagine what our software will be able to do for you.
But I am claiming that my 3dM approach makes the following possible:
- individually, images can be analysed by their shapes, patterns and components
- collectively, images can be sorted, ranked and selected depending on criteria that are set by experts in their field
- experts need to input their knowledge as ‘expert user’ in ‘expert portals’ where the help construct vocabularies and collectively agree on tolerance levels for values that the software finds.
In practice, automated image analysis will lead to:
- controlling the quality of drugs
- spotting differences in the quality of composite materials
- detecting upper and lower tolerance levels for standards at nanoscale and below.