Siemens Aktiengesellschaft

Deutsche Post Ag

Thomas Breuer

Bernd Hartmann

Gisbert Berger

Michael Zettler

Thomas Schoellhorn

J&uuml;rgen Franke

Andreas Marschner

Wilfried Hanisch

Thomas Schöllhorn

An image, which can be analyzed by a computer and shows at least one face of an item, is generated for an item that is to be transported at a predefined time. By automatic analysis of the image, the arrangement determines, for at least one of a plurality of predetermined optically detectable features, the value of the feature for the image of the item. The identification feature value vector generated is automatically compared with stored registration feature value vectors. A record for a particular item contains six registration feature value vectors and each of the six registration feature value vectors specifies a value for each predefined optically detectable feature for each one of six faces of the item. When the identification feature value vector matches a stored registration feature value vector with sufficient accuracy, a message is generated.

Method and arrangement for transporting cuboidal items

A method and an arrangement are presented for pattern recognition on the basis of statistics. According to the method, for an object to be recognized on the basis of a complete set of two-class or multiclass classifiers, the association with each target class of the class set is estimated with a numerical value that is produced by cascaded use of polynomial classifiers. According to the invention, on a learning sample in which all class patterns to be recognized are sufficiently represented, there is a selection, from all the two-class or multiclass classifiers by way of their estimation vector spectrum, of those two-class or multiclass classifiers with estimations contributing the most to minimize a scalar quantity calculated over the estimation vector spectrum and having high separating relevance. The selected two-class or multiclass classifiers are subsequently used to form, via an expanded learning sample, estimation vectors from which expanded characteristic vectors are produced by polynomial linking. An evaluation classifier is formed on the basis of said characteristic vectors for estimating all target classes.

Method and arrangement for pattern recognition on the basis of statistics

Growing community of inventors

Constance, Germany

Thomas Breuer