Robert Boschgmbh

Oliver Willers

Sebastian Sudholt

Michael Curcic

Sven Zinober

Ulrich Kunz

Stephanie Abrecht

Shervin Raafatnia

Christopher Huth

Konrad Groh

Christoph Gladisch

Matthias Woehrle

Christian Heinzemann

Jens Oehlerking

Michael Rittel

A computer-implemented method for evaluating an image classifier, in which a classifier output of the image classifier is provided for the actuation of an at least semi-autonomous robot. The evaluation method includes: ascertaining a first dataset including image data and annotations being assigned to the image data, the annotations including information about the scene imaged in the respective image and/or about image regions to be classified and/or about movement information of the robot; ascertaining regions of the scenes that are reachable by the robot based on the annotations; ascertaining relevance values for image regions to be classified by the image classifier; classifying the image data of the first image dataset with the aid of the image classifier; evaluating the image classifier based on image regions correctly classified by the image classifier and incorrectly classified image regions, as well as the calculated relevance values of the corresponding image regions.

Method and device for evaluating an image classifier

A method for training an artificial neural network, in particular a Bayesian neural network, by way of training data sets, having a step of adapting the parameters of the artificial neural network depending on a loss function, the loss function encompassing a first term that represents an estimate of a lower bound of the distances between the classifications of the training data sets by the artificial neural network and the expected classifications of the training data sets. The loss function further encompasses a second term that is configured in such a way that differences in the aleatoric uncertainty in the training data sets over different samples of the artificial neural network are regulated.

Method for training an artificial neural network, artificial neural network, use of an artificial neural network, and corresponding computer program, machine-readable memory medium, and corresponding apparatus

A method is indicated for determining a confidence value of an object of a class detected in an input image with the aid of a trained neural network, including: producing an activation signature for the class of the detected object using a plurality of output images of a layer of the neural network, the input image being provided to the input of the neural network; scaling the activation signature to the dimension of the input image; comparing an object portion of the scaled activation signature with an activation signature distribution of all objects of the same class of a training data set of the neural network in order to determine the confidence value.

Method for determining a confidence value of a detected object

A method is described for determining a confidence value for an object of a class determined by a neural network in an input image. The method includes: preparing an activation signature with the aid of a multiplicity of output images of a layer of the neural network for the class of the object, with the input image being provided to the input of the neural network; scaling the activation signature to the size of the input image; comparing an overlapping area portion of an area of the activation signature with an area of an object frame in relation to the area of the activation signature in order to determine the confidence value.

Method for determining a confidence value of an object of a class

A method for estimating a global uncertainty of output data of a computer implemented main neural network. The method includes determining a first measure quantifying to which extent the current input data of the main neural network is following the same distribution as the data, which was used for training the main neural network; generating a second measure quantifying the main neural network's certainty in its own prediction based on the input data; ascertaining a third measure, based on an estimation of class-discriminative features in the input data and a comparison of these features with a class activation probability distribution, especially wherein the class activation probability distribution was created based on estimated class-discriminative features during the training of the main neural network; and determining the global uncertainty based on at least two measures of the first, second and third measure.

Method for estimating a global uncertainty of a neural network

A method is provided for checking a sensor value of a MEMS sensor. In the process, an output signal of the MEMS sensor is detected and the sensor value is ascertained as a function of the output signal. In addition, frequency components of the output signal are examined and a determination is made as to whether the ascertained sensor value is reliable or unreliable as a function of the examination of the frequency components. If the sensor value is determined to be unreliable, the sensor value is discarded or provided with a lower weighting, or a warning it output relating to the unreliability of the sensor value or an item of information about the unreliability of the sensor value is stored.

Method for checking a sensor value of a MEMS sensor

A method for protecting a MEMS unit, in particular a MEMS sensor, against infrared investigations, at least one area of the MEMS unit being doped, the at least one doped area absorbing, reflecting or diffusely scattering more than 50%, in particular more than 90%, of an infrared light incident upon it.

Method for protecting a MEMS unit against infrared investigations and MEMS unit

An assembly made up of a micro-electro-mechanical system as physical unclonable function, which in reaction to a challenge, outputs a response in accordance with a mapping rule, and a controllable control element which is equipped, in accordance with a control command, to adjust an ambient parameter influencing the mapping rule.

Adjustable physical unclonable function

A method is provided for protecting a MEMS unit against infrared investigations, at least one layer being built into the structure of the MEMS unit or at least one layer being applied on a surface of the MEMS unit. The at least one layer absorbs, reflects or diffusely scatters more than 50%, in particular more than 90% of an infrared light incident upon it.

A method is provided for protecting a MEMS unit, in particular a MEMS sensor, against infrared investigations, a surface patterning being performed for at least one first area of a surface of the MEMS unit, the first area absorbing, reflecting or diffusely scattering more than 50%, in particular more than 90% of an infrared light incident upon it.

Growing community of inventors

Korb, Germany

Oliver Willers