Feature extraction system for identifying text within a table image

In a feature extraction system which creates a hierarchical tree structure corresponding to a block template of a document page, wherein the hierarchical tree structure includes a plurality of nodes each of which represents a block of image data in the block template of a document page and contains address data defining locations of the image data in the document page, a method of identifying table image data and locating text image data within the table, comprising the steps of:

• inputting image data of a document page;
• performing block selection processing so as to identify and separate the image data into blocks having the same image type;
• identifying blocks of image data which contain table image data;
• locating text blocks within the table image data;
• horizontally sorting each text block located in the table image data in relation to each other text block located in the table image data based on horizontal position information;
• vertically sorting each text block located in the table image data in relation to each other text block located in the table image data based on vertical position information;
• calculating rows and columns based on a result of the vertical and the horizontal sorting steps;
• assigning column and row address information to each text block in the table image data based on the calculating step;
• and
• storing, in a node representing the block of table image data in the hierarchical tree, table address information and storing, in respective nodes representing text blocks branching from the node representing the table image data, text block address information.

In a feature extraction system for analyzing image data of an input document image, an apparatus for identifying image data as a table image and for identifying text data within the table image, comprising:

• input means for inputting image data of a document page;
• block selection processing means for performing block selection processing on the image data, said block selection processing means identifying and separating the image data into blocks having the same image type;
• identifying means for identifying table image data based on the separated blocks of image data;
• identifying means for identifying text blocks within the table image data;
• horizontal sorting means for sorting of each text block located in the table image data in relation to each other text block located in the table image data based on horizontal position information;
• vertical sorting means for sorting of each text block located in the table image data in relation to each other text block located in the table image data based on vertical position information;
• calculating means for calculating rows and columns based on a result of the vertical and the horizontal sorting;
• address assigning means for assigning column and row address information to each text block in the table image data;
• and
• storing means for storing the assigned address information of each text block.

For a feature extraction system for analyzing image data of an input document image, a computer-readable memory medium storing computer-executable process steps to identify image data as a table image and to identify text within the table image, the steps comprising:

• an inputting step to input image data of a document page;
• a performing step to perform block selection processing on the image data, said block selection processing identifying and separating the image data into blocks having the same image type;
• an identifying step to identify table image data based on the separated blocks of image data;
• an identifying step to identifying text blocks within the table image data;
• a horizontally-sorting step to horizontally sort each text block located in the table image data in relation to each other text block located in the table image data based on horizontal position information;
• a vertically-sorting step to vertically sort each text block located in the table image data in relation to each other text block located in the table image data based on vertical position information;
• a calculating step to calculate rows and columns based on a result of the vertically and the horizontally-sorting steps;
• an assigning step to assign a column address and a row address to each text block in the table image data based on the calculating step;
• and
• a storing step to store the assigned addresses of each text block.

For a feature extraction system which creates a hierarchical tree structure corresponding to a block template of a document page, wherein the hierarchical tree structure includes a plurality of nodes each of which represents a block of image data in the block template of a document page and contains address data defining locations of the image data in the document page, a computer-readable memory medium storing computer-executable process steps to identify table image data and to locate text image data within the table, the steps comprising:

• an inputting step to input image data of a document page;
• a performing step to perform block selection processing so as to identify and separate the image data into blocks having the same image type;
• an identifying step to identify blocks of image data which contain table image data;
• a locating step to locate text blocks within the table image data;
• a horizontally-sorting step to horizontally sort each text block located in the table image data in relation to each other text block located in the table image data based on horizontal position information;
• a vertically-sorting step to vertically sort each text block located in the table image data in relation to each other text block located in the table image data based on vertical position information;
• a calculating step to calculate rows and columns based on a result of the vertical and the horizontal sorting steps;
• an assigning step to assign column and row address information to each text block in the table image data based on the calculating step;
• and
• a storing step to store, in a node representing the block of table image data in the hierarchical tree, table address information, and to store, in respective nodes representing text blocks branching from the node representing the table image data, text block address information.