公开(公告)号：US11429964B2

公开(公告)日：2022-08-30

申请号：US16711698

申请日：2019-12-12

Applicant: SAP SE

Inventor： Juliy Broyda ,  Lev Sigal ,  Anton Ioffe ,  Yuri Arshavski

IPC: G06Q20/40 ,  G06N20/00 ,  G06Q40/00 ,  G06K9/62 ,  G06N3/02 ,  G06F40/284 ,  G06Q20/04 ,  G06Q20/38 ,  G06T7/73 ,  G06N3/08 ,  G06T7/00 ,  G06V30/224 ,  G06V30/413 ,  G06V30/414 ,  G06V30/418 ,  G06F16/2455

Abstract: The present disclosure involves systems, software, and computer implemented methods for transaction auditing. One example method includes receiving a request to authenticate a document image. The image is preprocessed to prepare the image for line orientation analysis. The preprocessed image is analyzed to determine lines in the preprocessed image. The determined lines are automatically analyzed by performing line orientation test(s) on the determined lines to generate line orientation test result(s) for the preprocessed image. The line orientation test result(s) are evaluated to determine whether the image is authentic. In response to determining that at least one line orientation test result matches a predefined condition corresponding to an unauthentic document, a determination is made that the image is not authentic. In response to determining that none of the line orientation test results match any predefined condition corresponding to an unauthentic document, a determination is made that the image is authentic.

公开(公告)号：US20230162292A1

公开(公告)日：2023-05-25

申请号：US18149769

申请日：2023-01-04

Applicant: SAP SE

Inventor： Suchitra Sundararaman ,  Jesper Lind ,  Juliy Broyda ,  Lev Sigal ,  Anton Ioffe ,  Yuri Arshavski

IPC: G06Q40/12 ,  G06N20/00 ,  G06Q20/40 ,  G06N3/02 ,  G06F40/284 ,  G06Q20/04 ,  G06Q20/38 ,  G06T7/73 ,  G06T7/00 ,  G06V30/224 ,  G06V30/413 ,  G06V30/414 ,  G06V30/418 ,  G06N3/08 ,  G06F18/24 ,  G06V30/40

CPC classification number: G06Q40/12 ,  G06N20/00 ,  G06Q20/40 ,  G06Q20/4016 ,  G06N3/02 ,  G06F40/284 ,  G06Q20/045 ,  G06Q20/389 ,  G06T7/74 ,  G06T7/0002 ,  G06V30/224 ,  G06V30/413 ,  G06V30/414 ,  G06V30/418 ,  G06N3/08 ,  G06F18/24 ,  G06V30/40 ,  G06F16/24564

Abstract: The present disclosure involves systems, software, and computer implemented methods for transaction auditing. One example method includes training at least one machine learning model to determine features that can be used to determine whether an image is an authentic image of a document or an automatically generated document image, using a training set of authentic images and a training set of automatically generated document images. A request to classify an image as either an authentic image of a document or an automatically generated document image is received. The machine learning model(s) are used to classify the image as either an authentic image of a document or an automatically generated document image, based on features included in the image that are identified by the machine learning model(s). A classification of the image is provided. The machine learning model(s) are updated based on the image and the classification of the image.

公开(公告)号：US20210004580A1

公开(公告)日：2021-01-07

申请号：US16711642

申请日：2019-12-12

Applicant: SAP SE

Inventor： Suchitra Sundararaman ,  Jesper Lind ,  Juliy Broyda ,  Lev Sigal ,  Anton Ioffe ,  Yuri Arshavski

IPC: G06K9/00 ,  G06N20/00 ,  G06K9/62 ,  G06N3/02

Abstract: The present disclosure involves systems, software, and computer implemented methods for transaction auditing. One example method includes training at least one machine learning model to determine features that can be used to determine whether an image is an authentic image of a document or an automatically generated document image, using a training set of authentic images and a training set of automatically generated document images. A request to classify an image as either an authentic image of a document or an automatically generated document image is received. The machine learning model(s) are used to classify the image as either an authentic image of a document or an automatically generated document image, based on features included in the image that are identified by the machine learning model(s). A classification of the image is provided. The machine learning model(s) are updated based on the image and the classification of the image.

公开(公告)号：US12039615B2

公开(公告)日：2024-07-16

申请号：US18149769

申请日：2023-01-04

Applicant: SAP SE

Inventor： Suchitra Sundararaman ,  Jesper Lind ,  Juliy Broyda ,  Lev Sigal ,  Anton Ioffe ,  Yuri Arshavski

IPC: G06F18/24 ,  G06F40/284 ,  G06N3/02 ,  G06N3/08 ,  G06N20/00 ,  G06Q20/04 ,  G06Q20/38 ,  G06Q20/40 ,  G06Q40/12 ,  G06T7/00 ,  G06T7/73 ,  G06V30/224 ,  G06V30/40 ,  G06V30/413 ,  G06V30/414 ,  G06V30/418 ,  G06F16/2455

CPC classification number: G06Q40/12 ,  G06F18/24 ,  G06F40/284 ,  G06N3/02 ,  G06N3/08 ,  G06N20/00 ,  G06Q20/045 ,  G06Q20/389 ,  G06Q20/40 ,  G06Q20/4016 ,  G06T7/0002 ,  G06T7/74 ,  G06V30/224 ,  G06V30/40 ,  G06V30/413 ,  G06V30/414 ,  G06V30/418 ,  G06F16/24564 ,  G06T2207/20061 ,  G06T2207/30176

Abstract: The present disclosure involves systems, software, and computer implemented methods for transaction auditing. One example method includes training at least one machine learning model to determine features that can be used to determine whether an image is an authentic image of a document or an automatically generated document image, using a training set of authentic images and a training set of automatically generated document images. A request to classify an image as either an authentic image of a document or an automatically generated document image is received. The machine learning model(s) are used to classify the image as either an authentic image of a document or an automatically generated document image, based on features included in the image that are identified by the machine learning model(s). A classification of the image is provided. The machine learning model(s) are updated based on the image and the classification of the image.

公开(公告)号：US20210004949A1

公开(公告)日：2021-01-07

申请号：US16711698

申请日：2019-12-12

Applicant: SAP SE

Inventor： Juliy Broyda ,  Lev Sigal ,  Anton Ioffe ,  Yuri Arshavski

IPC: G06T7/00 ,  G06Q40/00 ,  G06N3/08 ,  G06K9/00 ,  G06T7/73

Abstract: The present disclosure involves systems, software, and computer implemented methods for transaction auditing. One example method includes receiving a request to authenticate a document image. The image is preprocessed to prepare the image for line orientation analysis. The preprocessed image is analyzed to determine lines in the preprocessed image. The determined lines are automatically analyzed by performing line orientation test(s) on the determined lines to generate line orientation test result(s) for the preprocessed image. The line orientation test result(s) are evaluated to determine whether the image is authentic. In response to determining that at least one line orientation test result matches a predefined condition corresponding to an unauthentic document, a determination is made that the image is not authentic. In response to determining that none of the line orientation test results match any predefined condition corresponding to an unauthentic document, a determination is made that the image is authentic.

公开(公告)号：US11568400B2

公开(公告)日：2023-01-31

申请号：US16711642

申请日：2019-12-12

Applicant: SAP SE

Inventor： Suchitra Sundararaman ,  Jesper Lind ,  Juliy Broyda ,  Lev Sigal ,  Anton Ioffe ,  Yuri Arshavski

IPC: G06K9/62 ,  G06Q20/40 ,  G06N20/00 ,  G06Q40/00 ,  G06N3/02 ,  G06F40/284 ,  G06Q20/04 ,  G06Q20/38 ,  G06T7/73 ,  G06T7/00 ,  G06V30/224 ,  G06V30/413 ,  G06V30/414 ,  G06V30/418 ,  G06N3/08 ,  G06F16/2455

Abstract: The present disclosure involves systems, software, and computer implemented methods for transaction auditing. One example method includes training at least one machine learning model to determine features that can be used to determine whether an image is an authentic image of a document or an automatically generated document image, using a training set of authentic images and a training set of automatically generated document images. A request to classify an image as either an authentic image of a document or an automatically generated document image is received. The machine learning model(s) are used to classify the image as either an authentic image of a document or an automatically generated document image, based on features included in the image that are identified by the machine learning model(s). A classification of the image is provided. The machine learning model(s) are updated based on the image and the classification of the image.

公开(公告)号：US09973789B1

公开(公告)日：2018-05-15

申请号：US15602747

申请日：2017-05-23

Applicant: SAP SE

Inventor： Ran Bittmann ,  Michael Kemelmakher ,  Yuri Arshavski

IPC: H04N21/234 ,  H04N21/81 ,  G06K9/00

CPC classification number: H04N21/23418 ,  G06K9/00718 ,  H04N21/812

Abstract: Methods, systems, and computer-readable storage media for receiving a set of frames, each frame being provided as a digital image that depicts a portion of an event and a logo associated with a brand, for each frame in the set of frames, and for each pixel in a frame: determining a weight of the pixel based on a distribution assigned to the frame, providing a quality of the logo depicted in the frame, and calculating a pixel quotient based on the weight and the quality, for each frame in the set of frames: determining a frame quotient at least partially based on a sum of all pixel quotients for the frame, and determining an impact indicator for the logo based on a total size of digital media comprising the set of frames and a sum of frame quotients of the frames in the set of frames.