公开(公告)号：US20210027113A1

公开(公告)日：2021-01-28

申请号：US16518728

申请日：2019-07-22

Applicant: Raytheon Company

Inventor： Jonathan Goldstein ,  Shane A. Zabel

IPC: G06K9/62

Abstract: Discussed herein are systems, devices, and methods for automatic target recognition based on a non-visible input image. A method can include providing, as input to a first machine learning (ML) model for object classification, pixel data of a non-visible image, the first ML model including an encoder from a second ML model, the second ML model trained to generate a visible image representation of an input non-visible image, and receiving, from the first ML model, data indicating one or more objects present in the non-visible image.

公开(公告)号：US11468266B2

公开(公告)日：2022-10-11

申请号：US16586465

申请日：2019-09-27

Applicant: Raytheon Company

Inventor： Jonathan Goldstein ,  Stephen J. Raif ,  Philip A. Sallee ,  Jeffrey S. Klein ,  Steven A. Israel ,  Franklin Tanner ,  Shane A. Zabel ,  James Talamonti ,  Lisa A. Mccoy

IPC: G06K9/00 ,  G06K9/62 ,  G06V10/24 ,  G06V30/24

Abstract: A machine receives a large image having large image dimensions that exceed memory threshold dimensions. The large image includes metadata. The machine adjusts an orientation and a scaling of the large image based on the metadata. The machine divides the large image into a plurality of image tiles, each image tile having tile dimensions smaller than or equal to the memory threshold dimensions. The machine provides the plurality of image tiles to an artificial neural network. The machine identifies, using the artificial neural network, at least a portion of the target in at least one image tile. The machine identifies the target in the large image based on at least the portion of the target being identified in at least one image tile.

公开(公告)号：US11373064B2

公开(公告)日：2022-06-28

申请号：US16518728

申请日：2019-07-22

Applicant: Raytheon Company

Inventor： Jonathan Goldstein ,  Shane A. Zabel

IPC: G06K9/00 ,  G06K9/62 ,  G01S13/90 ,  G01S15/89

Abstract: Discussed herein are systems, devices, and methods for automatic target recognition based on a non-visible input image. A method can include providing, as input to a first machine learning (ML) model for object classification, pixel data of a non-visible image, the first ML model including an encoder from a second ML model, the second ML model trained to generate a visible image representation of an input non-visible image, and receiving, from the first ML model, data indicating one or more objects present in the non-visible image.

公开(公告)号：US20210097344A1

公开(公告)日：2021-04-01

申请号：US16586465

申请日：2019-09-27

Applicant: Raytheon Company

Inventor： Jonathan Goldstein ,  Stephen J. Raif ,  Philip A. Sallee ,  Jeffrey S. Klein ,  Steven A. Israel ,  Franklin Tanner ,  Shane A. Zabel ,  James Talamonti ,  Lisa A. Mccoy

IPC: G06K9/62 ,  G06K9/32 ,  G06K9/68

Abstract: A machine receives a large image having large image dimensions that exceed memory threshold dimensions. The large image includes metadata. The machine adjusts an orientation and a scaling of the large image based on the metadata. The machine divides the large image into a plurality of image tiles, each image tile having tile dimensions smaller than or equal to the memory threshold dimensions. The machine provides the plurality of image tiles to an artificial neural network. The machine identifies, using the artificial neural network, at least a portion of the target in at least one image tile. The machine identifies the target in the large image based on at least the portion of the target being identified in at least one image tile.