# US 7,466,875 B1 · Method and system for determining the legibility of text in an image

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 7,466,875 B1
- Document ID: US7466875B1
- Status: Granted
- Kind: B1
- Application number: 10/790,256
- Filed: 2004-03-01
- Granted: 2008-12-16
- Earliest claimed: 2004-03-01
- Assignees: Amazon Technologies, Inc.
- Inventors: Hilliard Siegel; David Chick
- Web page: https://patentgrep.com/patent/US7466875B1

## CPC

- G06V 10/993: {Evaluation of the quality of the acquired pattern}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

Legibility of text in an image of a page is determined by comparing a measure of the text in the page image with a measure of the page image itself. In one aspect, a measure of the text in the page image may be the height of a line of text, while the measure derived from the page image may be the height of the page image. A text-to-page height ratio is determined and compared to one or more thresholds for determining legibility. In another aspect of the invention, a measure of the text in a page image is obtained by measuring the word density in the page image, while the measure derived from the page image comprises compressing the page image and determining the size of the compressed image file. Legibility is then determined by comparing the measure of word density with the compressed image file size.

## Claims

### Claim 1

A computer-implemented method for determining the legibility of text in an image of a page, the method comprising:
(a) obtaining an image of a page having text therein;
(b) performing text recognition on text in the page image and producing a measure of the text;
(c) comparing the measure of the text in the page image with a measure of the page image itself; and
(d) determining the legibility of the text in the page image based on the comparison of the measure of the text and the measure of the page image.

### Claim 2 (depends on 1)

The method of claim 1, wherein performing text recognition comprises identifying a spatial dimension of the text in the page image.

### Claim 3 (depends on 2)

The method of claim 2, wherein the measure of the text in the page image is a measure of text height in the identified spatial dimension, and the measure of the page image is a measure of page image height in the same spatial dimension.

### Claim 4 (depends on 3)

The method of claim 3, wherein comparing the measure of text height to page image height includes dividing the text height by the page image height to produce a text-to-page height ratio.

### Claim 5 (depends on 4)

The method of claim 4, wherein determining the legibility of the text includes comparing the text-to-page height ratio to a threshold.

### Claim 6 (depends on 5)

The method of claim 5, wherein the text is determined to be not legible if the text-to-page height ratio does not satisfy the threshold.

### Claim 7 (depends on 5)

The method of claim 5, wherein text recognition is performed on text in multiple lines in the page image and each of the multiple lines of text produces a text-to-page height ratio, wherein each of the multiple text-to-page height ratios are compared to the threshold, and the legibility of the text in the page image is determined based on a percentage of the text-to-page height ratios that satisfy the threshold.

### Claim 8 (depends on 5)

The method of claim 5, wherein the threshold is a first threshold, the method further comprising comparing the text-to-page height ratio to a second threshold if the text-to-page height ratio satisfies the first threshold.

### Claim 9 (depends on 8)

The method of claim 8, wherein the text is determined to be legible if the text-to-page height ratio satisfies the second threshold.

### Claim 10 (depends on 8)

The method of claim 8, wherein the text is determined to be possibly legible if the text-to-page height ratio satisfies the first threshold but not the second threshold.

### Claim 11 (depends on 8)

The method of claim 8, wherein text recognition is performed on text in multiple lines in the page image and each of the multiple lines of text produces a text-to-page height ratio, wherein if a determined percentage of the multiple text-to-page height ratios satisfy the first threshold, the text-to-page height ratios are compared to the second threshold, and if a determined percentage of the text-to-page height ratios satisfy the second threshold, the text is determined to be legible.

### Claim 12 (depends on 1)

The method of claim 1, further comprising storing the page image for display if the text in the page image is determined to be legible.

### Claim 13 (depends on 1)

The method of claim 1, wherein if the text in the page image is determined to be not legible, a higher resolution image of the page is obtained.

### Claim 14 (depends on 1)

The method of claim 1, wherein performing text recognition includes identifying and counting words that comprise the text in the page image, and wherein producing a measure of the text includes calculating a measure of word density in the page image based on the number of words in the page image and the spatial size of the page image.

### Claim 15 (depends on 14)

The method of claim 14, wherein the measure of the page image comprises:
(a) compressing the page image to form a compressed image file; and
(b) determining the file size of the compressed image file.

### Claim 16 (depends on 15)

The method of claim 15, wherein determining the legibility of the text includes comparing the measure of word density to the file size of the compressed image file.

### Claim 17

A computer-implemented method for displaying a page image based on a determined legibility of text in the page image, the method comprising:
(a) obtaining an image of a page at a base resolution;
(b) obtaining a measure of the text in the page image;
(c) comparing the measure of the text in the page image with a measure of the page image itself to determine the legibility of text in the page image; and
(d) if the text in the page image is determined to be legible, then displaying the page image, otherwise obtaining and displaying a new image of the page obtained at a resolution higher than the base resolution.

### Claim 18 (depends on 17)

The method of claim 17, wherein the base resolution page image is obtained from an original high resolution image of the page, and if the text in the base resolution page image is determined to be not legible, then the new page image is obtained from the original high resolution page image at a resolution higher than the base resolution.

### Claim 19 (depends on 17)

The method of claim 17, wherein obtaining the measure of text in the page image includes performing text recognition on text in the page image.

### Claim 20 (depends on 19)

The method of claim 19, wherein performing text recognition comprises identifying a spatial dimension of the text in the page image, wherein the measure of the text in the page image is a measure of text height in the identified spatial dimension, and wherein the measure of the page image is a measure of page image height in the same identified spatial dimension.

### Claim 21 (depends on 20)

The method of claim 20, wherein comparing the measure of text height to page image height includes dividing the text height by the page image height to produce a text-to-page height ratio, and wherein determining the legibility of the text includes comparing the text-to-page height ratio to a threshold.

### Claim 22 (depends on 19)

The method of claim 19, wherein text recognition is performed on multiple lines of text in the page image.

### Claim 23 (depends on 19)

The method of claim 19, wherein performing text recognition includes identifying and counting words that comprise the text in the page image, wherein producing the measure of the text includes calculating a measure of word density in the page image based on the number of words in the page image and the spatial size of the page image, wherein the measure of the page image includes compressing the page image to form a compressed image file and determining the file size of the compressed image file, and wherein determining the legibility of the text includes comparing the measure of word density to the file size of the compressed image file.

### Claim 24

A multistage method for automated determination of legibility of text in an image of a page, comprising:
(a) obtaining an image of a page having text therein;
(b) determining the legibility of the text by applying a first test of legibility to the text in the page image, wherein the first test of legibility includes obtaining a measure of the text in the page image and comparing the measure of the text with a measure of the page image itself;
(c) determining the legibility of the text by applying a second test of legibility to the text in the page image; and
(d) storing the page image for display if the text in the page image is determined to be legible.

### Claim 25 (depends on 24)

The method of claim 24, wherein the page image is stored for display if the text in the page image is determined to be legible by either the first test or the second test.

### Claim 26 (depends on 24)

The method of claim 24, wherein the page image is stored for display if the text in the page image is determined to be legible based on the results of both the first test and the second test.

### Claim 27 (depends on 24)

The method of claim 24, wherein if the text in the page image is determined not legible based on the first test, the second test is not applied to the page image.

### Claim 28 (depends on 24)

The method of claim 24, wherein the result of the first test is one of legible, not legible, or possibly legible, and wherein the second test is applied if the result of the first test is possibly legible.

### Claim 29

A computer-implemented system configured to determine the legibility of text in an image of a page, comprising a processor and a computer-readable medium containing computer-executable instructions that, when executed by the processor, cause the system to obtain an image of a page having text therein, perform text recognition on the text in the page image and produce therefrom a measure of the text, compare the measure of the text in the page image with a measure of the page image itself, and determine the legibility of the text in the page image based on the comparison of the measure of the text and the measure of the page image.

### Claim 30 (depends on 29)

The system of claim 29, wherein the computer-executable instructions, when executed by the processor, further cause the system to measure the text in the page image by measuring a height of a line of text, and further measure the height of the page image to derive the measure of the page image.

### Claim 31 (depends on 30)

The system of claim 30, wherein the computer-executable instructions, when executed by the processor, further cause the system to determine the legibility of the text by comparing the measure of text height with the page image height.

### Claim 32 (depends on 29)

The system of claim 29, further comprising a memory in communication with the processor for storing the page image for display if the text in the page image is determined to be legible.

### Claim 33 (depends on 29)

The system of claim 29, wherein if the text in the page image is determined to be not legible, the computer-executable instructions, when executed by the processor, further cause the system to obtain a higher resolution image of the page.

### Claim 34 (depends on 29)

The system of claim 29, wherein the computer-executable instructions, when executed by the processor, further cause the system to identify and count words that comprise the text in the page image and produce a measure of the text by calculating a word density based on the number of words in the page image and a spatial size of the page image.

### Claim 35 (depends on 34)

The system of claim 34, wherein the computer-executable instructions, when executed by the processor, further cause the system to determine the legibility of the text by compressing the page image to form a compressed image file, determine the file size of the compressed image file, and compare the calculated word density with the file size of the compressed image file.

### Claim 36 (depends on 29)

The system of claim 29, further comprising a display in communication with the processor for displaying the page image based on a determined legibility of the text in the page image.

### Claim 37 (depends on 36)

The system of claim 36, wherein if the text in the page image is determined to be not legible, the computer-executable instructions, when executed by the processor, further cause the system to obtain and communicate to the display a higher resolution image of the page.

### Claim 38 (depends on 29)

The system of claim 29, further comprising a memory in communication with the processor, wherein the computer-executable instructions, when executed by the processor, further cause the system to determine the legibility of the text by applying a first test of legibility and a second test of legibility to the text in the page image, and store the page image in the memory for display if the text in the page image is determined to be legible.

### Claim 39 (depends on 38)

The system of claim 38, wherein the computer-executable instructions, when executed by the processor, further cause the system to store the page image for display if the text in the page image is determined to be legible by either the first or the second test of legibility.

### Claim 40 (depends on 38)

The system of claim 38, wherein the computer-executable instructions, when executed by the processor, further cause the system to store the page image for display if the text in the page image is determined to be legible based on the results of both the first and second tests of legibility.

### Claim 41 (depends on 38)

The system of claim 38, wherein if the processor determines the text in the page image is not legible based on the first test of legibility, the computer-executable instructions, when executed by the processor, further cause the system to not apply the second test of legibility to the page image.
