Siemens Aktiengesellschaft

Baris Erol

Jason Dube

Dominic Trinko

Yuan Zi

Annemarie Breu

Philip Gregor

Joseph Tylka

A computer-implemented method for automatically orienting product containers includes acquiring a reference product label image associated with a batch of product containers on a production line and using a trained deep learning model to compute a perceptual center of the reference product label image based on perceptually relevant label features of the reference product label image. A production line camera acquires an image slice of the respective product label of individual product containers in the batch of product containers. A rotation angle for rotating an individual product container is computed based on the acquired image slice of the respective product label and a label center determined based on the computed perceptual center of the reference product label image. The computed rotation angle is communicated to a controller to effect a rotation of the individual product container.

System and method for automatically orienting product containers

A method for detecting defects in a formed metal part includes locating one or more regions of interest in a synthetic image of a part manufactured by a forming process. The synthetic image is informed based on a physics-based simulation of the forming process. The regions of interest indicate a high risk of having a defect from the forming process. A set of training images including real images of actual manufactured parts are registered with the synthetic image. The regions of interest are overlaid on each training image, to extract patches from the training images that correspond to high-risk regions. An anomaly detection model is trained on the patches extracted from the training images to detect a defect in a formed metal part from an acquired image of the formed metal part, by detecting an anomaly in a patch extracted from the acquired image that corresponds to a high-risk region.

Physics-informed anomaly detection in formed metal parts

An AI-based method for visual inspection of parts manufactured on a shop floor includes acquiring a set of real images of nominal parts manufactured on the shop floor to create training datasets. A self-supervised pre-trainer module is used to pre-train a loss computation neural network in a self-supervised learning process using a first dataset on pretexts defined by real-world conditions pertaining to the shop floor. The first dataset is labeled by automatically extracting pretext-related information from image metadata. A main anomaly trainer module is used to train a main anomaly detection neural network to reconstruct a nominal part image from an input manufactured part image in an unsupervised learning process using a second dataset. The main anomaly training measures a perceptual loss between an input image and a reconstructed image by measuring a difference between feature representations thereof at one or more layers of the pre-trained loss computation neural network.

Growing community of inventors

Rochester Hills, MI, United States of America

Baris Erol