Method and system that efficiently prepares text images for optical-character recognition
First Claim
1. An image-processing system that acquires an image containing text with curved text lines and generates a corresponding corrected image in which the text lines are straightened and have a rectilinear organization, the image-processing system comprising:
- one or more processors, one or more memories; and
computer instructions stored in one or more physical data-storage devices that, when executed by one or more of the one or more processors, control the image-processing system toreceive the text-containing image as an input,identify a page sub-image within the text-containing image,correct a vertical perspective of the page sub-image,generate contour lines to represent text lines of the page sub-image,generate a text-line-curvature model for the page sub-image in view of the contour lines, wherein the text-line-curvature model includes an inclination-angle map that associates each pixel in the page sub-image with an inclination angle of a respective contour line corresponding to a respective pixel, andstraighten the text lines within the received text-containing image using the text-line-curvature model.
4 Assignments
0 Petitions
Accused Products
Abstract
The current document is directed to methods and systems that straighten curvature in the text lines of text-containing digital images, including text-containing digital images generated from the two pages of an open book. Initial processing of a text-containing image identifies the outline of a text-containing page. Next, contours are generated to represent each text line. The midpoints and inclination angles of the links or vectors that comprise the contour lines are determined. A model is constructed for the perspective-induced curvature within the text image. In one implementation, the model, essentially an inclination-angle map, allows for assigning local displacements to pixels within the page image which are then used to straighten the text lines in the text image. In another implementation, the model is essentially a pixel-displacement map which is used to straighten the text lines in the text image.
50 Citations
21 Claims
-
1. An image-processing system that acquires an image containing text with curved text lines and generates a corresponding corrected image in which the text lines are straightened and have a rectilinear organization, the image-processing system comprising:
-
one or more processors, one or more memories; and computer instructions stored in one or more physical data-storage devices that, when executed by one or more of the one or more processors, control the image-processing system to receive the text-containing image as an input, identify a page sub-image within the text-containing image, correct a vertical perspective of the page sub-image, generate contour lines to represent text lines of the page sub-image, generate a text-line-curvature model for the page sub-image in view of the contour lines, wherein the text-line-curvature model includes an inclination-angle map that associates each pixel in the page sub-image with an inclination angle of a respective contour line corresponding to a respective pixel, and straighten the text lines within the received text-containing image using the text-line-curvature model. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15)
-
-
16. An image-processing system that acquires an image containing text with curved text lines and generates a corresponding corrected image in which the text lines are straightened and have a rectilinear organization, the image-processing system comprising:
-
one or more processors, one or more memories; and computer instructions stored in one or more physical data-storage devices that, when executed by one or more of the one or more processors, control the image-processing system to receive the text-containing image as an input, identify a page sub-image within the text-containing image, correct a vertical perspective of the page sub-image, generate contour lines to represent text lines of the page sub-image, generate a displacement map for the page sub-image in view of the contour lines, wherein; the displacement map associates each pixel in the page sub-image with a displacement of a respective pixel from a median height of a corresponding contour line, and the image-processing system generates the displacement map by; generating a displacement for each link of each contour line; initializing values of a set of coefficients for the displacement map, wherein the displacement map is a polynomial displacement model; and using a data-fitting method to determine coefficients for the polynomial displacement model from the displacements associated with the links, and straighten the text lines within the received text-containing image using the displacement map. - View Dependent Claims (17)
-
-
18. A method that receives a text-containing image and generates a corresponding corrected image in which the text lines are straightened and have a rectilinear organization, the method comprising:
-
receiving the text-containing image as an input, identifying a page sub-image within the text-containing image, correcting a vertical perspective of the page sub-image, generating contour lines to represent text lines of the page sub-image, generating a text-line-curvature model for the page sub-image in view of the contour lines, wherein the text-line-curvature model includes an inclination-angle map that associates each pixel in the page sub-image with an inclination angle of a respective contour line corresponding to a respective pixel, and straightening the text lines within the received text-containing image using the text-line-curvature model. - View Dependent Claims (19)
-
-
20. A method that receives a text-containing image and generates a corresponding corrected image in which the text lines are straightened and have a rectilinear organization, the method comprising:
-
receiving the text-containing image as an input, identifying a page sub-image within the text-containing image, correcting a vertical perspective of the page sub-image, generating contour lines to represent text lines of the page sub-image, generating a displacement map for the page sub-image in view of the contour lines, wherein; the displacement map that associates each pixel in the page sub-image with a displacement of a respective pixel from a median height of a corresponding contour line; and the generating of the displacement map further comprises; generating a displacement for each link of each contour line; initializing values of a set of coefficients for the displacement map, wherein the displacement map is a polynomial displacement model; and using a data-fitting method to determine coefficients for the polynomial displacement model from the displacements associated with the links, and straightening the text lines within the received text-containing image using the displacement map.
-
-
21. A non-transitory computer-readable medium having recorded thereon instructions, that when executed by one or more computer processors for image-processing system, perform operations comprising:
-
receiving a text-containing image as an input, identifying a page sub-image within the text-containing image, correcting a vertical perspective of the page sub-image, generating contour lines to represent text lines of the page sub-image, generating a text-line-curvature model for the page sub-image in view of the contour lines, wherein the text-line-curvature model includes an inclination-angle map that associates each pixel in the page sub-image with an inclination angle of a respective contour line corresponding to a respective pixel, and straightening the text lines within the received text-containing image using the text-line-curvature model.
-
Specification