G06V 10/764

Definition

(Für diese Definition ist die deutsche Übersetzung noch nicht abgeschlossen)

Classification of images or videos to identify the category or set of categories (classes) to which a new observation belongs, on the basis of a training set of data containing observations (or instances) whose category membership is known.

Novelty detection (e.g. classification of “unseen” observations), anomaly detection or outlier detection.

Notes – technical background

These notes provide more information about the technical subject matter that is classified in this place:

Individual observations may be analysed into a set of quantifiable properties, known as explanatory variables or features. These properties may be categorical, ordinal, integer-valued, real-valued, etc. Other classifiers perform a class assignment by comparing current observations to previous observations by means of a similarity or distance function.

A classifier can be parametric or non-parametric depending on the type of model adopted for the observations.

Classification algorithms include those:

• based on the distance between a decision surface and training patterns, e.g. support vector machines (SVM);
• based on the distance between the pattern to be recognised and a reference, where the reference can be a prototype, a centroid of samples of the same class or the closest patterns from the same class or different classes, e.g. nearest-neighbour classification;
• based on a parametric, probabilistic model, where the model uses the Neyman-Pearson lemma, likelihood ratios, receiver operating characteristics (ROC), plotting the false acceptance rate (FAR) versus the false rejection rate (FRR), Bayesian classification, etc.;
• based on a graph-like or tree-like model, e.g. decision trees, random forests, etc. Examples are the classification and regression trees (CART), ID3 (Iterative Dichotomiser 3), C4.5, etc.

The decision surface of the classifier may be a linear classifier or a non-linear classifier. Linear classifiers model the boundaries between different classes in the feature space as hyperplanes. Non-linear classifiers use e.g. quadratic, polynomial, or hyperbolic functions instead.

Examples

A linear support vector machine classifier which attempts to define a linear boundary between two classes (205, 210) of feature vectors originating from images containing “persons” and “non-persons”, such as to separate them into two different classes.

Decision tree classifying objects in the image data using an efficient hardware implementation with FIFO buffers

Querverweise

Informative Querverweise

Glossar