DEVICES AND METHODS FOR RESTORING LOW-RESOLUTION TEXT IMAGES
First Claim
1. A system that extracts text from an image, comprising:
- a capture device that captures the image having a low resolution;
an image segmentation subsystem that partitions the image into image segments;
an image restoration subsystem that generates a resolution-expanded image from the image segments and that negates degradation effects of the low-resolution image by transforming the image segments from a first domain to a second domain and deconvolving the transformed image segments in the second domain to determine parameters of the low-resolution image; and
a text recognition subsystem that transforms the restored image data into computer readable text data based on the determined parameters.
2 Assignments
0 Petitions
Accused Products
Abstract
A system that extracts text from an image includes a capture device that captures the image having a low resolution. An image segmentation subsystem partitions the image into image segments. An image restoration subsystem generates a resolution-expanded image from the image segments and negates degradation effects of the low-resolution image by transforming the image segments from a first domain to a second domain and deconvolving the transformed image segments in the second domain to determine parameters of the low-resolution image. A text recognition subsystem transforms the restored image data into computer readable text data based on the determined parameters.
-
Citations
38 Claims
-
1. A system that extracts text from an image, comprising:
-
a capture device that captures the image having a low resolution; an image segmentation subsystem that partitions the image into image segments; an image restoration subsystem that generates a resolution-expanded image from the image segments and that negates degradation effects of the low-resolution image by transforming the image segments from a first domain to a second domain and deconvolving the transformed image segments in the second domain to determine parameters of the low-resolution image; and a text recognition subsystem that transforms the restored image data into computer readable text data based on the determined parameters. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16)
-
-
17. A system that expands the resolution of a text image, comprising:
-
a text segmentation subsystem that partitions the text image into characters; and a restoration subsystem comprising; a first transform subsystem that transforms the partitioned text image from an image domain into a Radon domain; an interpolator that linearly interpolates each column of the Radon domain image; a first deconvolution subsystem that deconvolves each column to estimate first parameters of the text image; an image expander that forms an expanded Radon domain image; a second transform subsystem that transforms the expanded Radon domain image into the image domain; an image slicer that slices the expanded image domain image to enforce a first constraint; a third transform subsystem that transforms the sliced image from the image domain into the Radon domain; a second deconvolution subsystem that estimates second parameters of the text image in the Radon domain; a fourth transform subsystem that transforms the expanded Radon domain image including the second parameters into the image domain; and an output that outputs the expanded image domain image as a restored text image. - View Dependent Claims (18, 19, 20, 21)
-
-
22. A method of extracting text from an image, comprising:
-
capturing the image having a low resolution; partitioning the image into image segments; and generating a restored, resolution-expanded image from the image segments and negating degradation effects of the low-resolution image by transforming the image segments from a first domain to a second domain and by deconvolving the transformed image segments in the second domain to determine parameters of the low-resolution image; and transforming the restored image data into computer readable text data based on the determined parameters. - View Dependent Claims (23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)
-
-
35. A method of expanding the resolution of a text image, comprising:
-
partitioning the text image into characters; transforming the partitioned text image from an image domain into a Radon domain; linearly interpolating each column of the Radon domain image; deconvolving each column by applying first determined parameters of the text image;
forming an expanded Radon domain image;transforming the expanded Radon domain image into the image domain; slicing the expanded image to enforce a first constraint; transforming the sliced image from the image domain to the Radon domain; estimating second parameters of the text image in the Radon domain; transforming the expanded Radon domain image including the second parameters into the image domain; and outputting the expanded image domain image as a restored text image. - View Dependent Claims (36, 37, 38)
-
Specification