A system for extraction and verification of handwritten signatures from arbitrary documents. The system comprises one or more computing devices configured to: receive a digital image of a document; perform a dilating transformation via convolution matrix on the digital image to obtain a modified image; determine a plurality of regions of connected markings in the digital image; based at least in part on a pixel density or proximity to an anchor substring of each region, determine whether any region contains any handwritten signature; extract first image data of the region containing a handwritten signature from the digital image; retrieve second image data of a confirmed example signature for a purported signer of the handwritten signature; and based on a comparison of the first image data with the second image data, forward a determination of whether the first image data and second image data are similar.

A method and a system for extracting information from a table in a document is provided. The method includes: receiving a document that includes information that is arranged in a table; determining three sets of coordinates that respectively relate to lines, words, and characters included in the document; extracting a list of lines based on the first set of coordinates; reconstructing the rows of the table based on list of lines and the second set of coordinates; reconstructing the columns of the table based on the reconstructed rows and the third set of coordinates; and outputting a reconstruction of the table. The three sets of coordinates are expressible in an hOCR format that is based on an open standard for representation of scanned information that is obtainable by using an optical character recognition (OCR) technique.

Systems and methods for training a model are described herein. In one example, a system for training the model includes a processor and a memory in communication with the processor having a training module. The training module has instructions that cause the processor to determine a contrastive loss using a self-supervised contrastive loss function, adjust, based on the contrastive loss, model weights a visual backbone that generated feature maps and/or a textual backbone that generated feature vectors. The training module also has instructions that cause the processor to determine a localized loss using a supervised loss function that compares an image-caption attention map with visual identifiers and adjust, based on the localized loss, the model weights the visual backbone and/or the textual backbone.

A method, a computer program product, and a computer system compare images for content consistency. The method includes receiving a first image including a first document and a second image including a second document. The method includes performing a visual classification analysis on the first image and the second image. The visual classification analysis generates an overlap of the first image with the second image. The method includes determining whether a region of the overlap is indicative of a content inconsistency. As a result of the region of the overlap being indicative of a content inconsistency, the method includes performing a character recognition analysis on a first area of the first image and a second area of the second image corresponding to the region of the overlap to verify the content inconsistency.

The system may perform a method comprising analyzing metadata of a text layer of a page of a first pdf document to determine that the pdf document is a first true pdf document; receiving the first true pdf document, in response to the first pdf document being the first true pdf document; receiving a selection of a field including first data to be extracted from the first true pdf document; displaying the first data; creating a template including the coordinates corresponding to the selected field and the first data of the first true pdf document; and extracting from an accessible text layer of a second true pdf document, second data based on the template from the first true pdf document.

A system trains a machine learning model which recognizes characters of text images. The system stores the machine learning model which recognizes characters of text images. The machine learning model includes a character segmentation network which is configured to extract visual features from text images, and to generate character bounding boxes from the text images, a domain adaptation network configured to classify the text images into domains based on the visual features, and a text recognition network configured to recognize characters in the text images based on the character bounding boxes and the visual features. The system is configured to (1) reverse gradients in the training of the domain adaptation network to minus gradients and back-propagate the minus gradients through the character segmentation network (2) back-propagate gradients in the training of the text recognition network through the character segmentation network.

A system trains a machine learning model which recognizes characters of text images. The system stores the machine learning model which recognizes characters of text images. The machine learning model includes a character segmentation network which is configured to extract visual features from text images, and to generate character bounding boxes from the text images, a domain adaptation network configured to classify the text images into domains based on the visual features, and a text recognition network configured to recognize characters in the text images based on the character bounding boxes and the visual features. The system is configured to (1) reverse gradients in the training of the domain adaptation network to minus gradients and back-propagate the minus gradients through the character segmentation network (2) back-propagate gradients in the training of the text recognition network through the character segmentation network.

Users of a database management engine may generate fillable digital documents by mapping interface elements onto form documents. When a user maps interface elements onto a form document, the user may accidentally overlap two or more interface elements. To rectify this, the database management engine may modify the position of one of interface elements based on a set of positioning rules. In addition, the database management engine may identify and suggest mappings to users based on similar documents that have been previously mapped. The database management engine identifies similar documents using information about the document, the user, and the mapping itself. The mapping associated with the most similar document may be provided to the user as a suggested mapping. The database management engine converts the form document and finalized mapping into a fillable digital document. The fillable digital document is sent to recipients, who complete the fillable digital document.

There are provided a learning recommendation apparatus and method for detecting a problem from an image through character recognition and providing at least one sub-topic learning among a plurality of sub-topic learnings related to the detected problem. The provided learning recommendation apparatus recommends, as a recommendation target, a plurality of learning topics including the concept of a formula which has been read through the character recognition for an image, wherein a priority order is set to the plurality of learning topics based on the concept distance between the learning topic and the learning history, and the learning topics are recommended so that the learning topic having a higher priority order is located at a higher position.

A display control integrated circuit (IC) applicable to performing real-time video content text detection and speech automatic generation in a display device may include a pre-processing circuit, a character recognition circuit and a post-processing circuit. The pre-processing circuit may input a video signal to obtain a real-time video content carried by the video signal, and perform preliminary text detection on the real-time video content to generate a series of segmented character images to indicate a subtitle. The character recognition circuit may perform character recognition on the series of segmented character images to generate a series of characters, respectively. The post-processing circuit may perform vocabulary correction on the series of characters to selectively replace any erroneous character with a correct character to generate one or more vocabularies, for performing speech automatic generation.