Automatic delineation and extraction of tabular data using machine learning

A system comprising:

• a memory;
• and
• a processor coupled with the memory, the processor programmed with machine learning algorithms and configured to perform a method comprising:
• training a machine learning model, the training comprising:
• receiving a set of images of tabular data and a set of markup data corresponding respectively to the images of tabular data;
• training a first neural network to delineate the tabular data from the set of images into cells using the markup data, wherein training the first neural network comprises parsing a first image from the set of images to match a portion of the tabular data from the first image as specified by a corresponding first markup data from the set of markup data, and updating one or more weights from the first neural network based on a structural accuracy computed by comparing the portion with the corresponding first markup data;
• and
• training a second neural network to determine content of the cells in the tabular data from the set of images using the markup data;
• receiving an input image containing a first tabular data without markup data indicative of a structure of the first tabular data;
• and
• generating an electronic output corresponding to the first tabular data by determining the structure of the first tabular data using the first neural network and extracting content of the first tabular data using the second neural network.

A computer program product comprising a memory storage device having computer executable instructions stored thereon, the computer executable instructions when executed by a processor cause the processor to perform a method comprising:

• training a machine learning model, the training comprising:
• receiving a set of images of tabular data and a set of markup data corresponding respectively to the images of tabular data;
• training a first neural network to delineate the tabular data from the set of images into cells using the markup data, wherein training the first neural network comprises parsing a first image from the set of images to match a portion of the tabular data from the first image as specified by a corresponding first markup data from the set of markup data, and updating one or more weights from the first neural network based on a structural accuracy computed by comparing the portion with the corresponding first markup data;
• and
• training a second neural network to determine content of the cells in the tabular data from the set of images using the markup data;
• receiving an input image containing a first tabular data without markup data indicative of a structure of the first tabular data;
• and
• generating an electronic output corresponding to the first tabular data by determining the structure of the first tabular data using the first neural network and extracting content of the first tabular data using the second neural network.