Processing

Please wait...

Settings

Settings

Goto Application

1. WO2022042940 - METHOD AND DEVICE FOR CLASSIFYING TARGETS

Publication Number WO/2022/042940
Publication Date 03.03.2022
International Application No. PCT/EP2021/070009
International Filing Date 16.07.2021
IPC
G06K 9/00 2006.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
G06K 9/46 2006.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
36Image preprocessing, i.e. processing the image information without deciding about the identity of the image
46Extraction of features or characteristics of the image
G06K 9/62 2006.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
CPC
G06K 9/0053
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
00496Recognising patterns in signals and combinations thereof
00523Feature extraction
0053by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
G06K 9/6271
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
6267Classification techniques
6268relating to the classification paradigm, e.g. parametric or non-parametric approaches
627based on distances between the pattern to be recognised and training or reference patterns
6271based on distances to prototypes
G06V 10/50
G06V 20/56
Applicants
  • IBEO AUTOMOTIVE SYSTEMS GMBH [DE]/[DE]
Inventors
  • NITSCH, Julia
  • FELLENBERG, Christian
  • POSEWSKY, Thorbjörn
  • ERDMANN, Jennifer
  • HOFSÄSS, Cornelia
Agents
  • RGTH PATENTANWÄLTE PARTGMBB
Priority Data
20193251.428.08.2020EP
Publication Language German (de)
Filing Language German (DE)
Designated States
Title
(DE) VERFAHREN UND VORRICHTUNG ZUR KLASSIFIZIERUNG VON TARGETS
(EN) METHOD AND DEVICE FOR CLASSIFYING TARGETS
(FR) PROCÉDÉ ET DISPOSITIF POUR CLASSIFIER DES CIBLES
Abstract
(DE) Es wird ein Verfahren (100) zur Klassifizierung von Targets vorgeschlagen, das das Extrahieren (107) von Merkmalen aus Messdaten eines oder mehrerer Empfangselemente eines Sensors (11) mittels eines neuronalen Netzwerkes (14a) oder mittels eines Gaussian Mixture Models (GMM; 14b) umfasst, wobei es sich bei den Messdaten des mindestens einen Empfangselementes des Sensors (11) jeweils um zumindest einen Abschnitt eines Photonhistogramms (40) handelt, und wobei es sich bei dem neuronalen Netzwerk (14a) um ein fully connected neuronales Netzwerk (FCN) oder ein convolutional neuronales Network (CNN) handelt.
(EN) The present invention relates to a method (100) for classifying targets, which comprises the extraction (107) of features from measurement data of one or more receiving elements of a sensor (11) using a neural network (14a) or a Gaussian Mixture Model (GMM; 14b), wherein the measurement data of the at least one receiving element of the sensor (11) is in each case at least a portion of a photon histogram (40), and wherein the neural network (14a) is a fully connected neural network (FCN) or a convolutional neural network (CNN).
(FR) La présente invention concerne un procédé (100) pour classifier des cibles, lequel procédé comprend l'extraction (107) de caractéristiques à partir de données de mesure d'un ou plusieurs éléments de réception d'un capteur (11) à l'aide d'un réseau neuronal (14a) ou d'un modèle de mélange gaussien (GMM; 14b), les données de mesure de l'au moins un élément de réception du capteur (11) étant, dans chaque cas, au moins une partie d'un histogramme de photons (40), et le réseau neuronal (14a) étant un réseau neuronal entièrement connecté (FCN) ou un réseau neuronal à convolution (CNN).
Related patent documents
Latest bibliographic data on file with the International Bureau