Method and system for text-image orientation
View Patent ↗The current application is directed to a method and system for automatically determining the sense orientation of regions of scanned-document images that include symbols and characters of languages that are not written as simple sequential strings of alphabetic characters. In one implementation, the sense-orientation method and system to which the current application is directed employs a relatively small set of orientation-marker characters that occur frequently in printed text and that lack rotational symmetry. In this implementation, text-character images within a region of a scanned-document image are compared to each of a set of orientation-marker patterns corresponding to orientation-marker characters in order to identify images corresponding to orientation-marker patterns in the text-containing region of the scanned-document image and to determine an overall sense orientation for the text-containing region of the scanned-document image based on the orientations of the identified orientation-marker patterns.
1. An image processing system that determines an orientation for a text-containing region of an image, the image processing system comprising:
one or more processors;
one or more electronic memories;
one or more mass-storage devices;
an orientation-marker pattern set stored in one or more of the one or more electronic data storage devices, wherein the orientation-marker pattern set comprises two or more patterns corresponding to two or more considered orientations of one or more orientation-marker characters; and
computer instructions stored in the one or more electronic data storage devices that control the image processing system to
receive an image with a text-containing region,
store a representation of the text-containing region in one or more of the one or more electronic memories,
identify one or more text-character images within the text-containing region that match an orientation-marker pattern from the orientation-marker pattern set, and
determine an orientation for the text-containing region from the orientations of the one or more text-character images that match the orientation-marker pattern.
2. The image processing system of claim 1 wherein the number of orientation-marker patterns in the orientation-marker-pattern set is less than 10% of the total number of patterns used to represent text symbols.
3. The image processing system of claim 1 wherein the number of orientation-marker patterns in the orientation-marker-pattern set is less than 1% of the total number of patterns used to represent text symbols.
4. The image processing system of claim 1 wherein the number of orientation-marker patterns in the orientation-marker-pattern set is less than 0.1% of the total number of patterns used to represent text symbols.
5. The image processing system of claim 1 wherein the orientation-marker is characters are asymmetric and a frequency of occurrence in text greater than a threshold frequency.
6. The image processing system of claim 1 wherein identifying the one or more text-character images within the text-containing region that match the orientation-marker pattern from the orientation-marker pattern set further comprises:
for a second number of text-character images within the text-containing region,
computing a number j of metric values for j different metrics with respect to the text-character image;
for one or more orientation-marker patterns of the orientation-marker pattern set,
determining a match score with respect to the text-character image, and
when the determined match score is less than a threshold value and when the match score is smaller than the smallest match score previously determined with respect to the text-character image, storing the match score as the smallest match score and a reference to the orientation-marker pattern as the best matching pattern.
7. The image processing system of claim 6 wherein determining a match score with respect to the text-character image further comprises:
determining a sum of weighted distances between the j computed metric values and j corresponding metric values for the orientation-marker pattern.
8. The image processing system of claim 7 wherein each of the j metrics is a function that is computed from a bit-map representation of either an orientation-marker pattern or a text-character image.
9. The image processing system of claim 1 wherein determining an orientation for the text-containing region from the orientations of the one or more text-character images that match the orientation-marker pattern further comprises:
for the one or more text-character images, determining an orientation associated with the text-character image as an orientation of the orientation-marker pattern that matches the text-character image; and
selecting, as the determined orientation, an orientation associated with a greatest number of the text-character images.
10. A method carried out by an image-processing system having
one or more processors,
one or more electronic memories,
one or more mass-storage devices; and
an orientation-marker pattern set stored in one or more of the one or more electronic data storage devices, wherein the orientation-marker pattern set comprises two or more patterns corresponding to two or more considered orientations of one or more orientation-marker characters,
the method determining orientation of a text-containing region of an image, the method comprising:
receiving an image with a text-containing region;
storing a representation of the text-containing region in one or more of the one or more electronic memories;
identifying one or more text-character images within the text-containing region that match an orientation-marker pattern from the orientation-marker pattern set, and
determining an orientation for the text-containing region from the orientations of the one or more text-character images that match the orientation-marker pattern.
11. The method of claim 10 wherein the number of orientation-marker patterns in the orientation-marker-pattern set is less than 0.1%, 1%, or 10% of the total number of patterns used to represent text symbols.
12. The method of claim 11 wherein the orientation-marker characters are asymmetric and have a frequency of occurrence in text greater than a threshold frequency.
13. The method of claim 11 wherein identifying the one or more text-character images within the text-containing region that match the orientation-marker pattern from the orientation-marker pattern set further comprises:
for a second number of text-character images within the text-containing region,
computing a number j of metric values for j different metrics with respect to the text-character image;
for one or more orientation-marker patterns of the orientation-marker-pattem set,
determining a match score with respect to the text-character image, and
when the determined match score is less than a threshold value and when the match score is smaller than the smallest match score previously determined with respect to the text-character image, storing the match score as the smallest match score and a reference to the orientation-marker pattern as the best matching pattern.
14. The method of claim 13 wherein determining a match score with respect to the text-character image further comprises:
determining a sum of weighted distances between the j computed metric values and j corresponding metric values for the orientation-marker pattern.
15. The method of claim 14 wherein each of the j metrics is a function that is computed from a bit-map representation of either an orientation-marker pattern or a text-character image.
16. The method of claim 11 wherein determine an orientation for the text-containing region from the orientations of the one or more text-character images that match the orientation-marker pattern further comprises:
selecting, as the determined orientation, an orientation most frequently associated with the orientation-marker patterns that match the text-character images.
17. An image processing system that determines an orientation for a text-containing region of an image, the image processing system comprising:
one or more processors;
one or more electronic memories;
one or more mass-storage devices;
an orientation-marker pattern set stored in one or more of the one or more electronic data storage devices, wherein the orientation-marker pattern set comprises two or more patterns corresponding to two or more considered orientations of one or more orientation-marker characters; and
computer instructions stored in the one or more electronic data storage devices that control the image processing system to determine an orientation for the text-containing region from orientations of one or more text-character images within the text-containing region that match an orientation-marker pattern from the orientation-marker pattern set.
18. The image processing system of claim 17 wherein the number of orientation-marker patterns in the orientation-marker-pattern set is less than 10% of the total number of patterns used to represent text symbols.
19. The image processing system of claim 1 wherein the number of orientation-marker patterns in the orientation-marker-pattern set is less than 1% of the total number of patterns used to represent text symbols.
20. The image processing system of claim 17 wherein the number of orientation-marker patterns in the orientation-marker-pattern set is less than 0.1% of the total number of patterns used to represent text symbols.