Automated method for alignment of document objects

• US 8,271,871 B2
• Filed: 04/30/2009
• Issued: 09/18/2012
• Est. Priority Date: 04/30/2009
• Status: Expired due to Fees

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Marchesotti, Luca
• Primary Examiner(s): Queler, Adam M
• Assistant Examiner(s): JOHNSON, AARON M
• Application Number:
  US12/432,948
• Publication Number:
  US 20100281361A1
• Time in Patent Office: 1,237 Days
• Field of Search: 715/243, 715/256, 715/235, 715/857, 715/253, 345/418, 382/298
• US Class Current: 715/243
• CPC Class Codes: G06F 40/103 Formatting, i.e. changing of presentation of documents automatic justification G06F40/189; automatic line break hyphenation G06F40/191
• CPC Class Codes: G06F 40/186 Templates
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/18143 Extracting features based on salient regional features, e.g. scale invariant feature transform [SIFT] keypoints
• CPC Class Codes: G06V 30/413 Classification of content, e.g. text, photographs or tables

Abstract

An apparatus and automated method are disclosed for alignment of objects in a document which allows saliency within one or both objects to be a factor in the alignment. The method includes, for an input electronic document, identifying first and second objects to be aligned on a page of the document. A one dimensional guideline profile is generated for at least the first object based on a detection of saliency for the first object. The first and second objects are aligned based on the guideline profile to form a modified document and the modified document is output.

Claims

1. A method for alignment of objects in a document comprising:

for an input electronic document, identifying first and second objects to be aligned on a page of the document;

detecting saliency for the first object, including generating a saliency map of the first object;

generating a first one-dimensional guideline profile for the first object based on the detection of saliency for the first object, the guideline profile being based on one dimensional averaging of saliency values in the saliency map;

generating a second one-dimensional guideline profile for the second object based on a detection of saliency for the second object;

aligning the first and second objects based on the first and second guideline profiles to form a modified document; and

outputting the modified document.

2. The method of claim 1, further comprising assigning at least the first object to one of a plurality of object classes, each class being associated with a saliency detection method, the plurality of classes including a first class which is associated with a first saliency detection method and a second class, which is associated with a second detection method.

3. The method of claim 2, wherein the assigning of at least the first object to an object class comprises assigning the first and second objects to a respective object class.

4. The method of claim 2, wherein the assignment includes using a classifier trained to classify objects according to class.

5. The method of claim 2, wherein the plurality of classes comprise at least one of:

a pictorial object class;

a text object class; and

a graphical element class.

6. The method of claim 5, wherein when one of the first and second objects is assigned to a pictorial object class, the pictorial object is categorized into one of at least three categories, a first of the categories being a People category, and a second of the categories being a Non-People category, and a third of the categories being a Sky category.

7. The method of claim 5, wherein when one of the first and second objects is assigned to a pictorial object class, the pictorial object is categorized into one of at least two categories, a first of the categories being a People category, and a second of the categories being a Non-People category, wherein when the pictorial object is categorized into the People category, the detection method includes detecting a location of at least one of a person'"'"'s face and a person'"'"'s eyes, the guideline profile having a peak value at a location corresponding to the person'"'"'s face or person'"'"'s eyes.

8. The method of claim 7, wherein when the pictorial object is categorized into the Non-People category, the detection method includes generating a saliency map.

9. The method of claim 2, wherein first class is a pictorial object class and the second class is a text object class.

10. The method of claim 1, wherein the aligning includes convoluting the first and second guideline profiles to identify a peak and aligning the objects based on a location of the peak.

11. The method of claim 10, wherein the aligning includes aligning a peak of the first guideline profile with a peak of the second guideline profile.

12. The method of claim 1, wherein one of the first and second objects is a fixed object and the other of the first and second objects is a floating object and the alignment includes moving the floating object.

13. The method of claim 1, wherein the first object comprises a pictorial object and the second object comprises a text object.

14. A non-transitory computer program product encoding instructions which, when executed by a computer causes the computer to execute the method of claim 1.

15. A computer apparatus comprising:

memory which stores instructions for performing the method of claim 1; and

a processor in communication with the memory for executing the instructions.

16. A method for alignment of objects in a document comprising:

for an input electronic document, identifying first and second objects to be aligned on a page of the document;

generating a first one-dimensional guideline profile for the first object based on a detection of saliency for the first object, wherein the first object has horizontal and vertical dimensions and the first one dimensional guideline profile has, as its one dimension, the vertical dimension or the horizontal dimension of the object;

generating a second one-dimensional guideline profile for the second object based on a detection of saliency for the second object;

aligning the first and second objects based on the first and second guideline profiles to form a modified document; and

outputting the modified document.

17. An apparatus for alignment of objects in a document comprising:

computer readable memory which stores an object alignment system comprising;

an object detector for identifying first and second objects to be aligned within a document;

a profile generator for generating a one dimensional guideline profile for at least the first object based on a detection of saliency for the first object, the detection of saliency including generating a saliency map of the first object, the guideline profile being based on one dimensional averaging of saliency values in the saliency map; and

an alignment generator for generating an alignment of the first and second objects based on the guideline profile; and

a processor in communication with the memory which implements the object alignment system.

18. The apparatus of claim 17, wherein the object alignment system further comprises a saliency detector selector which selects a saliency detection component based on a type of the first object, the profile generator generating the one dimensional guideline based on an output of the selected saliency detection component.

19. A computer implemented method for alignment of objects in a document comprising:

identifying objects to be aligned in an electronic document;

assigning an object class to each of the objects selected from a plurality of object classes, the classes including at least one pictorial object class and a text object class;

for each object to be aligned, selecting a respective saliency detection method based on the assigned object class;

applying the selected detection method for each object to be aligned to generate a saliency representation;

with a computer processor, generating a one dimensional guideline profile for each object to be aligned based on one dimensional averaging of saliency values in the saliency representation; and

aligning at least first and second of the objects based on the respective one dimensional guideline profiles to form a modified document, including translating one of the first and second objects relative to the other of the first and second objects; and

outputting the modified document.

20. The method of claim 19, wherein the generating of the one dimensional guideline profile for at least one object to be aligned is based on a combination of at least two saliency representations corresponding to different saliency detection methods.