Document image analysis apparatus, document image analysis method and program thereof
AssigneeRakuten Group, Inc.
InventorsSimona Maggio, Alois De La Comble, Ken Prepin
Life of this patent
Estimating the term…
Abstract
Disclosed herein is a document image analysis apparatus including: a document image acquisition unit configured to acquire a document image; a region detection unit configured to detect a plurality of regions from the document image acquired by the document image acquisition unit; a clustering unit configured to cluster the plurality of regions detected by the region detection unit to integrate into a cluster; and a reading order assignment unit configured to assign a reading order to a plurality of regions belonging to the cluster within the cluster integrated by the clustering unit.
Claims20
A document image analysis apparatus, comprising:
- at least one memory configured to store program code;
- and
- electronic circuitry including at least one of a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), and at least one processor, the at least one processor being configured to read and operate according to the program code, the electronic circuitry configured to:
- acquire a document image having an article, the article having at least a title and a column;
- detect a plurality of regions from the document image acquired, the plurality of regions representing respective portions of the article;
- classify the plurality of regions into any of a plurality of classes;
- cluster the plurality of regions detected to integrate into a cluster that corresponds to a single article;
- classify regions belonging to a single cluster into any of a plurality of super-classes based on the classified classes of the regions, wherein the plurality of super-classes comprises a title region, a column region and an illustration region;
- sort the classified regions within each of the plurality of super-classes and across the plurality of super-classes;
- and
- assign, within a single cluster, a reading order to at least some of the sorted classified regions, the reading order being an order in which the plurality of regions are to be read in the article.
A document image analysis method executed by a document image analysis apparatus, comprising:
- acquiring a document image having an article, the article having at least a title and a column;
- detecting a plurality of regions from the document image acquired, the plurality of regions representing respective portions of the article;
- classifying the plurality of regions into any of a plurality of classes;
- clustering the plurality of regions detected to integrate into a cluster that corresponds to a single article;
- classifying regions belonging to a single cluster into any of a plurality of super-classes based on the classified classes of the regions, wherein the plurality of super-classes comprises a title region, a column region and an illustration region;
- sorting the classified regions within each of the plurality of super-classes and across the plurality of super-classes;
- and
- assigning, within a single cluster, a reading order to at least some of the sorted classified regions, the reading order being an order in which the plurality of regions are to be read in the article.
A non-transitory computer readable medium having stored thereon a document image analysis computer program for causing a computer to execute document image analysis processing, the computer program causing the computer to:
- acquire a document image having an article, the article having at least a title and a column;
- detect a plurality of regions from the document image acquired, the plurality of regions representing respective portions of the article;
- classify the plurality of regions into any of a plurality of classes;
- cluster the plurality of regions detected to integrate into a cluster that corresponds to a single article;
- classify regions belonging to a single cluster into any of a plurality of super-classes based on the classified classes of the regions, wherein the plurality of super-classes comprises a title region, a column region and an illustration region;
- sort the classified regions within each of the plurality of super-classes and across the plurality of super-classes;
- and
- assign, within a single cluster, a reading order to at least some of the sorted classified regions, the reading order being an order in which the plurality of regions are to be read in the article.
Dependencies are inferred from the claim text.