LEARNING A FORM STRUCTURE
    2.
    发明公开

    公开(公告)号:US20240177514A1

    公开(公告)日:2024-05-30

    申请号:US18071465

    申请日:2022-11-29

    摘要: A system learns the structure of a form. The structure of the form can be learned from a single image (e.g., a photograph that includes the form) without user annotation. The form includes typewritten and handwritten text entries. The system groups text entries in the form based on lines detected in the form. The system then measures a distance and an angle between two text entry locations in the group of text entries. The group of text entries, the distances, and the angles can be captured in a bipartite graph. The bipartite graph represents possible pairing solutions where a typewritten text entry is paired with a handwritten text entry. The system identifies an optimal pairing solution, from the possible pairing solutions, using the distances and angles. The optimal pairing solution is identified by minimizing the standard deviation of the distances and/or by minimizing the circular standard deviation of the angles.

    HANDWRITTEN POSTAGE
    3.
    发明公开
    HANDWRITTEN POSTAGE 审中-公开

    公开(公告)号:US20230215221A1

    公开(公告)日:2023-07-06

    申请号:US18150687

    申请日:2023-01-05

    摘要: The technology described herein provides a handwritten postage that comprises handwriting on a postal item that forms a unique identifier for the postal item (e.g., envelope, postcard, sticker) when analyzed by a computer vision application. The unique identifier is computer derived from the handwritten postage and allows one instance of handwritten postage to be differentiated from all other instances of handwritten postage. The unique identifier may be derived from an image of an envelope that includes an instance of handwritten postage when the handwritten postage is activated. The unique identifier may be formed from a combination of handwriting content (e.g., to and from address), metadata (e.g., date activated), pre-printed content on the postal item (e.g., fiducial marks), post-printed content (e.g., to or from address) and the visual image created by all or a portion of the handwriting. Postage value is added to the handwritten postage through an activation process.

    Handwritten text line wrapping
    4.
    发明授权

    公开(公告)号:US11941349B1

    公开(公告)日:2024-03-26

    申请号:US17942531

    申请日:2022-09-12

    摘要: A computer-implemented method for handwritten text line wrapping includes: obtaining, from a user, at least two words of handwritten text on a screen; determining an original bounding box for the at least two words; creating at least one line-break character for the at least two words; determining at least one baseline for the at least two words; determining a new bounding box for the at least two words based on the at least one baseline; generating, on the screen, a text box; moving, on the screen, at least one of the at least two words from a first line of at least one line of handwritten text to a second line of the at least one line of handwritten text, wherein the second line of handwritten text fits within the text box; and adjusting at least one gap between the at least one line of handwritten text.

    AUTOMATIC GENERATION OF TRAINING DATA FOR HAND-PRINTED TEXT RECOGNITION

    公开(公告)号:US20230206674A1

    公开(公告)日:2023-06-29

    申请号:US17562344

    申请日:2021-12-27

    发明人: Jason James Grams

    摘要: A method for generating training data for hand-printed text recognition includes obtaining a structured document, obtaining a set of hand-printed character images and database metadata from a database, generating a modified document page image, and outputting a training file. The structured document includes a document page image that includes text characters and document metadata that associates each of the text characters to a document character label. The database metadata associates each of the set of hand-printed character images to a database character label. The modified document page image is generated by iteratively processing each of the text characters. The iterative processing includes determining whether an individual text character should be replaced, selecting a replacement hand-printed character image from the set of hand-printed character images, scaling the replacement hand-printed character image, and inserting the replacement hand-printed character image into the modified document page image.

    HANDWRITTEN TEXT LINE WRAPPING
    7.
    发明公开

    公开(公告)号:US20240086623A1

    公开(公告)日:2024-03-14

    申请号:US17942531

    申请日:2022-09-12

    摘要: A computer-implemented method for handwritten text line wrapping includes: obtaining, from a user, at least two words of handwritten text on a screen; determining an original bounding box for the at least two words; creating at least one line-break character for the at least two words; determining at least one baseline for the at least two words; determining a new bounding box for the at least two words based on the at least one baseline; generating, on the screen, a text box; moving, on the screen, at least one of the at least two words from a first line of at least one line of handwritten text to a second line of the at least one line of handwritten text, wherein the second line of handwritten text fits within the text box; and adjusting at least one gap between the at least one line of handwritten text.

    Automatic generation of training data for hand-printed text recognition

    公开(公告)号:US11715317B1

    公开(公告)日:2023-08-01

    申请号:US17562344

    申请日:2021-12-27

    发明人: Jason James Grams

    摘要: A method for generating training data for hand-printed text recognition includes obtaining a structured document, obtaining a set of hand-printed character images and database metadata from a database, generating a modified document page image, and outputting a training file. The structured document includes a document page image that includes text characters and document metadata that associates each of the text characters to a document character label. The database metadata associates each of the set of hand-printed character images to a database character label. The modified document page image is generated by iteratively processing each of the text characters. The iterative processing includes determining whether an individual text character should be replaced, selecting a replacement hand-printed character image from the set of hand-printed character images, scaling the replacement hand-printed character image, and inserting the replacement hand-printed character image into the modified document page image.