公开(公告)号：US10546242B2

公开(公告)日：2020-01-28

申请号：US15495313

申请日：2017-04-24

Applicant: General Electric Company

Inventor： Arpit Jain ,  Swaminathan Sankaranarayanan ,  David Scott Diwinsky ,  Ser Nam Lim ,  Kari Thompson

IPC: G06N7/00 ,  G06K9/62 ,  G06N3/08

Abstract: A method includes determining object class probabilities of pixels in a first input image by examining the first input image in a forward propagation direction through layers of artificial neurons of an artificial neural network. The object class probabilities indicate likelihoods that the pixels represent different types of objects in the first input image. The method also includes selecting, for each of two or more of the pixels, an object class represented by the pixel by comparing the object class probabilities of the pixels with each other, determining an error associated with the object class that is selected for each pixel of the two or more pixels, determining one or more image perturbations by back-propagating the errors associated with the object classes selected for the pixels of the first input image through the layers of the neural network without modifying the neural network, and modifying a second input image by applying the one or more image perturbations to one or more of the first input image or the second input image prior to providing the second input image to the neural network for examination by the neurons in the neural network for automated object recognition in the second input image.

公开(公告)号：US20180253866A1

公开(公告)日：2018-09-06

申请号：US15495313

申请日：2017-04-24

Applicant: General Electric Company

Inventor： Arpit Jain ,  Swaminathan Sankaranarayanan ,  David Scott Diwinsky ,  Ser Nam Lim ,  Kari Thompson

IPC: G06T7/00 ,  G06K9/62 ,  G06N3/04 ,  G06N3/08 ,  G06N7/00

CPC classification number: G06N7/005 ,  G06K9/6267 ,  G06K9/628 ,  G06N3/0454 ,  G06N3/0472 ,  G06N3/0481 ,  G06N3/084 ,  G06N3/088

Abstract: A method includes determining object class probabilities of pixels in a first input image by examining the first input image in a forward propagation direction through layers of artificial neurons of an artificial neural network. The object class probabilities indicate likelihoods that the pixels represent different types of objects in the first input image. The method also includes selecting, for each of two or more of the pixels, an object class represented by the pixel by comparing the object class probabilities of the pixels with each other, determining an error associated with the object class that is selected for each pixel of the two or more pixels, determining one or more image perturbations by back-propagating the errors associated with the object classes selected for the pixels of the first input image through the layers of the neural network without modifying the neural network, and modifying a second input image by applying the one or more image perturbations to one or more of the first input image or the second input image prior to providing the second input image to the neural network for examination by the neurons in the neural network for automated object recognition in the second input image.