IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

**Definition:** This place covers:

Higher-level interpretation and recognition of images or videos, which includes pattern recognition, pattern learning and semantic interpretation as fundamental aspects. These aspects involve the detection, categorisation, identification, authentication of image or video patterns. For this purpose, image or video data are acquired and preprocessed. In the next step, distinctive features are extracted. Based on these features or representations derived from them, matching, clustering or classification is performed which may lead to one or several decisions, related confidence values (e.g. probabilities), classification or clustering labels. The aim is to find an explanation or to derive a specific meaning.

Pattern recognition or pattern learning in a specific, image or video-related context that includes:

Further details are given in the Definition statement of group G06V10/00 . Image or video recognition can be carried out by using electronic means ( G06V10/70 ) or by using optical means ( G06V10/88 ).

Typically, a pattern recognition system involves one or more of the following techniques:

**Glossary:** - authentication: verifying the identity of a sample using a test of genuineness by undertaking a one-to-one comparison with the genuine (authentic) sample - categorisation: assigning a sample to a class according to certain distinguishing properties (or characteristics) of that class; it generally involves a one-to-many test in which one data sample is compared with the characteristics of several classes. - classification: assigning labels to patterns - clustering: grouping or separating samples in groups or classes according to their (dis)similarity or closeness. It generally involves many-to-many comparisons using a (dis)similarity measure or a distance function. - feature extraction: deriving descriptive or quantitative measures from data. - identification: in the context of collecting of samples, identification means selecting a particular sample having a (predefined) characteristic which distinguishes it from the others. Several samples are generally matched against the one to be identified in a many-to-one process. - image and video understanding: techniques for semantic interpretation, pattern recognition or pattern learning specifically applied to images and videos - pattern: data having characteristic regularity, or a representation derived from it, having some explanatory value or a meaning, e.g. an object depicted in an image