Grid Layout Determination from a Document Image

A computing device implemented to determine a grid layout from a document image, the computing device comprising:

• a memory to maintain the document image of the document, as well as the grid layout of rows and columns in the document;
• a processor system configured to execute a grid layout system implemented to perform operations comprising to:
• determine feature elements of the document content in the document;
• generate a node tree of bounded elements that represent relationships of the feature elements in the document, each of the bounded elements being considered in the determination of the grid layout of the rows and columns in the document;
• generate a containment model that includes the node tree of the bounded elements, and includes size and position data for each of the bounded elements in the document;
• and
• determine a column layout of the columns in the document and a row layout of the rows in the document based on the containment model.

A system for determining a grid layout from a document image, the system comprising:

• means for storing the document image of the document;
• means for determining the grid layout of rows and columns in the document, the means for determining the grid layout comprising means for:
• determining feature elements of the document content in the document;
• generating a node tree of bounded elements that represent relationships of the feature elements in the document, each of the bounded elements being considered in the determination of the grid layout of the rows and columns in the document;
• generating a containment model that includes the node tree of the bounded elements, the containment model representing an organization of the bounded elements in the node tree;
• and
• determining a column layout of the columns in the document and a row layout of the rows in the document based on the containment model.