知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

6 records (took 0.014 seconds)

    Classification with segmentation neural network for image-based content capture
    1.
    发明授权
    Classification with segmentation neural network for image-based content capture 有权

    公开(公告)号：US10977524B2

    公开(公告)日：2021-04-13

    申请号：US16381962

    申请日：2019-04-11

    Applicant: Open Text SA ULC

    Inventor： Sreelatha Reddy Samala

    IPC: G06K9/00 ,  G06K9/62 ,  G06T5/20 ,  G06T7/11

    Abstract: A segmentation neural network is extended to provide classification at the segment level. An input image of a document is received and processed, utilizing a segmentation neural network, to detect pixels having a signature feature type. A signature heatmap of the input image can be generated based on the pixels in the input image having the signature feature type. The segmentation neural network is extended from here to further process the signature heatmap by morphing it to include noise surrounding an object of interest. This creates a signature region that can have no defined shape or size. The morphed heatmap acts as a mask so that each signature region or object in the input image can be detected as a segment. Based on this segment-level detection, the input image is classified. The classification result can be provided as feedback to a machine learning framework to refine training.

    Classification with segmentation neural network for image-based content capture
    3.
    发明授权
    Classification with segmentation neural network for image-based content capture 有权

    公开(公告)号：US12211244B2

    公开(公告)日：2025-01-28

    申请号：US18325127

    申请日：2023-05-30

    Applicant: Open Text SA ULC

    Inventor： Sreelatha Reddy Samala

    IPC: G06T5/00 ,  G06F18/21 ,  G06F18/2413 ,  G06T5/20 ,  G06T7/11 ,  G06V10/26 ,  G06V10/82 ,  G06V30/19 ,  G06V30/413

    Abstract: A segmentation neural network is extended to provide classification at the segment level. An input image of a document is received and processed, utilizing a segmentation neural network, to detect pixels having a signature feature type. A signature heatmap of the input image can be generated based on the pixels in the input image having the signature feature type. The segmentation neural network is extended from here to further process the signature heatmap by morphing it to include noise surrounding an object of interest. This creates a signature region that can have no defined shape or size. The morphed heatmap acts as a mask so that each signature region or object in the input image can be detected as a segment. Based on this segment-level detection, the input image is classified. The classification result can be provided as feedback to a machine learning framework to refine training.

    CLASSIFICATION WITH SEGMENTATION NEURAL NETWORK FOR IMAGE-BASED CONTENT CAPTURE
    4.
    发明申请
    CLASSIFICATION WITH SEGMENTATION NEURAL NETWORK FOR IMAGE-BASED CONTENT CAPTURE 有权

    公开(公告)号：US20210224577A1

    公开(公告)日：2021-07-22

    申请号：US17222940

    申请日：2021-04-05

    Applicant: Open Text SA ULC

    Inventor： Sreelatha Reddy Samala

    IPC: G06K9/62 ,  G06T5/20 ,  G06T7/11

    Abstract: A segmentation neural network is extended to provide classification at the segment level. An input image of a document is received and processed, utilizing a segmentation neural network, to detect pixels having a signature feature type. A signature heatmap of the input image can be generated based on the pixels in the input image having the signature feature type. The segmentation neural network is extended from here to further process the signature heatmap by morphing it to include noise surrounding an object of interest. This creates a signature region that can have no defined shape or size. The morphed heatmap acts as a mask so that each signature region or object in the input image can be detected as a segment. Based on this segment-level detection, the input image is classified. The classification result can be provided as feedback to a machine learning framework to refine training.

    CLASSIFICATION WITH SEGMENTATION NEURAL NETWORK FOR IMAGE-BASED CONTENT CAPTURE
    6.
    发明申请
    CLASSIFICATION WITH SEGMENTATION NEURAL NETWORK FOR IMAGE-BASED CONTENT CAPTURE 审中-公开

    公开(公告)号：US20200327360A1

    公开(公告)日：2020-10-15

    申请号：US16381962

    申请日：2019-04-11

    Applicant: Open Text SA ULC

    Inventor： Sreelatha Reddy Samala

    IPC: G06K9/62 ,  G06T5/20 ,  G06T7/11

    Abstract: A segmentation neural network is extended to provide classification at the segment level. An input image of a document is received and processed, utilizing a segmentation neural network, to detect pixels having a signature feature type. A signature heatmap of the input image can be generated based on the pixels in the input image having the signature feature type. The segmentation neural network is extended from here to further process the signature heatmap by morphing it to include noise surrounding an object of interest. This creates a signature region that can have no defined shape or size. The morphed heatmap acts as a mask so that each signature region or object in the input image can be detected as a segment. Based on this segment-level detection, the input image is classified. The classification result can be provided as feedback to a machine learning framework to refine training.

Patent Agency Ranking