公开(公告)号：US11663815B2

公开(公告)日：2023-05-30

申请号：US17191338

申请日：2021-03-03

Applicant: Infosys Limited

Inventor： Ujwal Bhate ,  Ninad Kulkarni ,  Shaurya Dwivedi

IPC: G06T7/00 ,  G06K9/62 ,  G06V10/94

CPC classification number: G06T7/0008 ,  G06K9/6257 ,  G06K9/6262 ,  G06K9/6269 ,  G06V10/95 ,  G06V10/955 ,  G06T2207/10016 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30136 ,  G06V2201/06

Abstract: Examples of the present invention provides a method and system for inspection of heat recovery steam generator (HRSG) equipment to identify defects and damages using computer vision and deep learning techniques. The method comprising capturing one or more input frames by one or more input devices, classifying the one or more input frames by a scenario classifier to identify a scenario type based on a first modelled data prepared by training one or more deep neural networks (DNN), selecting at least one damage detector based on the identified scenario type, identifying one or more damage types by the at least one damage detector based on second modelled data prepared by training the one or more DNN and displaying one or more output frame indicating the identified one or more damage types of the HRSG equipment.

公开(公告)号：US20210304400A1

公开(公告)日：2021-09-30

申请号：US17191338

申请日：2021-03-03

Applicant: Infosys Limited

Inventor： Ujwal Bhate ,  Ninad Kulkarni ,  Shaurya Dwivedi

IPC: G06T7/00 ,  G06K9/62 ,  G06K9/00

Abstract: Examples of the present invention provides a method and system for inspection of heat recovery steam generator (HRSG) equipment to identify defects and damages using computer vision and deep learning techniques. The method comprising capturing one or more input frames by one or more input devices, classifying the one or more input frames by a scenario classifier to identify a scenario type based on a first modelled data prepared by training one or more deep neural networks (DNN), selecting at least one damage detector based on the identified scenario type, identifying one or more damage types by the at least one damage detector based on second modelled data prepared by training the one or more DNN and displaying one or more output frame indicating the identified one or more damage types of the HRSG equipment.

公开(公告)号：US12131501B2

公开(公告)日：2024-10-29

申请号：US17394175

申请日：2021-08-04

Applicant: Infosys Limited

Inventor： Ujwal Bhate ,  Ninad Jayant Kulkarni ,  Mohammedshadab Mohammedaslam Shaikh ,  Arnab Chakravarty ,  Pratyush Choubey ,  Neeraj Kumar Gulia ,  Jigar Sanjay Shah

IPC: G06T7/73 ,  G06N3/045 ,  G06T7/11 ,  G06T7/20 ,  G06T17/20

CPC classification number: G06T7/75 ,  G06N3/045 ,  G06T7/11 ,  G06T7/20 ,  G06T17/20 ,  G06T2207/20084 ,  G06T2207/30184

Abstract: A method and/or system for automated estimation of 3D orientation of a physical asset using deep learning models and computer vision algorithms, according to one or more embodiments. The system may be configured to receive images of the physical asset and camera orientation data as input, use deep learning neural network models to isolate the physical assets across the images, track each physical asset instance throughout the images and derive a 3D point cloud model of each asset by projecting binary masks of the asset contours from different view-points. The 3D point cloud model is further processed and supplemented with camera orientation data to estimate the 3D orientation of one or more assets present in the images.

公开(公告)号：US20220309708A1

公开(公告)日：2022-09-29

申请号：US17394175

申请日：2021-08-04

Applicant: Infosys Limited

Inventor： Ujwal Bhate ,  Ninad Jayant Kulkarni ,  Mohammedshadab Mohammedaslam Shaikh ,  Arnab Chakravarty ,  Pratyush Choubey ,  Neeraj Kumar Gulia ,  Jigar Sanjay Shah

IPC: G06T7/73 ,  G06N3/04 ,  G06T7/11 ,  G06T7/20 ,  G06T17/20

Abstract: A method and/or system for automated estimation of 3D orientation of a physical asset using deep learning models and computer vision algorithms, according to one or more embodiments. The system may be configured to receive images of the physical asset and camera orientation data as input, use deep learning neural network models to isolate the physical assets across the images, track each physical asset instance throughout the images and derive a 3D point cloud model of each asset by projecting binary masks of the asset contours from different view-points. The 3D point cloud model is further processed and supplemented with camera orientation data to estimate the 3D orientation of one or more assets present in the images.