Introduction to computerized image analysis

Interpretation

Interpretation of the image includes matching and recognition. In matching we start from an image A and an image with the same content as the original image or as part of it B. We try to find where in A B fits better. It can be that B is an image of a face and that we want to find that specific face in an image of a number of people - A. For B to "fit" in A it can be necessary to scale or rotate the image.

A typical example of recognition is document analysis, i.e., recognition of characters. This is what for instance the Post Office is using to sort the mail. A machine reads the address written on the envelope and decides from that where to send the mail. All this is automatically done.