Automated system for indexing graphical documents having associated text labels

• US 5,845,288 A
• Filed: 12/11/1995
• Issued: 12/01/1998
• Est. Priority Date: 12/11/1995
• Status: Expired due to Term

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Syeda-Mahmood, Tanveer F.
• Primary Examiner(s): Von Buhr, Maria N.
• Application Number:
  US08/570,022
• Time in Patent Office: 1,086 Days
• Field of Search: 395/613, 395/615, 395/611, 395/603, 395/777, 395/762, 395/773, 395/133, 707/3, 707/100, 707/102, 707/104, 707/502, 707/512-515, 382/176, 382/180
• US Class Current: 1/1
• CPC Class Codes: G06F 16/58 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/5846 using extracted text
• CPC Class Codes: G06F 16/5854 using shape and object relationship
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

A system for image indexing for the domain of technical manual documents depicting line drawing images of physical equipment. Such line drawings are often associated with text labels that indicate references to component parts in an associated table. The indexing system locates regions containing a machine part in relevant pages of the manual in response to a query describing the part. The query supplied by the user is a textual description of the machine part from which an associated geometric description of the object is retrieved. The indexing mechanism includes two stages, i.e., selection and recognition. The selection phase involves isolating relevant images from the document database using a combination to textual and visual processing and indicating the appropriate regions within those images that are likely to contain the queried machine part. Model-based object recognition then confirms the presence of the part at that location by finding a match of features between the part and the model using a 3D model description associated with the textual query.

Claims

1. An automated indexing system for indexing documents, which include graphics having associated text labels, in a manner allowing object querying within said documents, comprising:

a microprocessor for;

a) processing graphic portions of a document and identifying isolated graphic objects within said graphic portions;

b) recognizing text labels in said graphic portions;

c) relating said isolated graphic objects to text labels associated therewith; and

a database for storing references to said isolated graphic objects and said text labels associated therewith in a queryable database index of said database, for independent retrieval of said isolated graphic objects and said associated text labels upon directing an independent query thereto.

2. A method of processing a document containing graphics and text for use in an indexing system associated with a computer implemented database management system, wherein:

a bitmapped image containing graphic portions further comprised of isolated graphic objects having associated text labels is received and stored;

said bitmapped image is processed to identify said isolated graphic objects and to identify and recognize said text labels therein;

said text labels and said isolated graphic objects are associated;

references to said isolated graphic objects and said associated text labels are stored in a queryable database index, for retrieval of graphic objects and associated text upon directing a query independently thereto.

3. The method as described in claim 2, wherein said processing of said bitmapped image to identify said isolated graphic objects therein includes identifying groups of graphic objects connected to said isolated objects within said image.

4. The method as described in claim 2, wherein said association of said text labels and said isolated graphic objects includes identifying lead lines associated with said text labels, each lead line having a text end and a pointing end;

and

identifying isolated graphic objects near said pointing end of said lead lines, whereby said isolated graphic objects near said pointing end of said lead lines are associated with said text characters at the end of said lead line.

5. An automated image indexing system for indexing graphical objects within graphical portions of documents in a manner allowing querying for said objects independently of said graphical portions, comprising:

means for processing graphic portions of documents into isolated connected graphical regions;

means for approximately localizing queried graphical objects by associating existing annotating text words with said connected graphical regions; and

means for automatically isolating single graphical objects from said connected graphics regions.

6. The system of claim 5 further comprising a means for verifying a match to posed queries using visual object recognition by matching object features from said posed queries with stored geometric models representing said objects.

7. The system of claim 5 further comprising a means for recognizing annotating text words in bit-mapped line-drawing images.

8. The system of claim 7 further comprising a means for forming data representations in a database management system using feature maps extracted from said line-drawing images.

9. The system of claim 8 further comprising a means for verifying a match to posed queries using visual object recognition by matching object features from said posed queries with stored geometric models representing said objects.

10. The system of claim 7 further comprising a means for verifying a match to posed queries using visual object recognition by matching object features from said posed queries with stored geometric models representing said objects.

11. The system of claim 7 wherein said graphical documents have associated text tables which are used for indexing said graphical documents.

12. The system of claim 11 further comprising means for receiving queries and suitably transforming them for indexing.

13. The system of claim 5 further comprising a means for recognizing annotating text words in bit-mapped line-drawing images.

14. The system of claim 13 further comprising means for receiving queries and suitably transforming them for indexing.

15. The system of claim 14 further comprising a means for verifying a match to posed queries using visual object recognition by matching object features from said pose queries with stored geometric models representing said objects.

16. The system of claim 5 wherein said graphical portions of said documents have associated text tables related to said graphical, said system further comprising:

means for receiving queries and transforming them for indexing;

means for recognizing annotating text words in bit-mapped line-drawing images;

means for forming data representations in a database management system using feature maps extracted from said line-drawing images; and

means for verifying a match to posed queries using visual object recognition by matching, using said feature maps, object features from said posed queries with stored geometric models representing said objects.