System and method for measuring mobile document image quality

Patent metadata

Field	Value
Publication number	US-2016188993-A1
Application number	US-201414586285-A
Country	US
Kind code	A1
Filing date	Dec 30, 2014
Priority date	Dec 30, 2014
Publication date	Jun 30, 2016
Grant date	—

Abstract

A system and method for detecting the quality of a captured digital image depicting a hardcopy document are disclosed. The captured digital image is analyzed to determine a corresponding blurriness, noise, hotspot, and uneven illumination metric representing a quality level of the image data. The blurriness, noise, hotspot, and uneven illumination metrics are then combined to formulate a pass/caution/fail indicator for the user to respond to the captured digital image quality.

First claim

1 . A system for measuring image quality comprising:
a digital imaging system configured to capture an image; a data storage system configured to store the captured image as a gray scale image; a processing system configured to access the stored gray scale image, wherein the processor system includes several processing modules, including:
a hotspot detector configured to detect the presence of a hotspot in the image and to set a hotspot flag as true if a hotspot is detected;
a noise detector configured to detect the presence of noise in the image and to set a noise flag as true if noise is detected;
an uneven illumination detector for detecting the presence of uneven illumination in the image and to set an uneven illumination flag as true if uneven illumination is detected; and
a blurriness detector configured to detect the presence of blurriness in the image and to set a blurriness flag as true if blurriness is detected; and
an output indicator configured to notify a user if any flags were set as true by the hotspot detector, noise detector, uneven illumination detector, or blurriness detector.

2 . The system of claim 1 , wherein the image further comprises text.

3 . The system of claim 1 , wherein the image further comprises a color image, and wherein the processing system is further configured to convert the image into the grey scale image.

4 . The system of claim 1 , wherein said hotspot detector module is configured to:
determine a minimum value of the pixels in the stored gray scale image; determine a maximum value of the pixels in the stored gray scale image; determine a mean value for the stored gray scale image; determine a standard deviation for the stored gray scale image; and evaluate the maximum value, minimum value, mean value, and standard deviation to determine whether a hotspot is present.

5 . The system of claim 1 , wherein the noise detector module is configured to:
apply a median blur to an image plane of the stored gray scale image to create a signal plane; subtract a signal plane from the image plane to create a residual noise plane; determine signal and noise values for the image; and determine whether there is noise present in the image.

6 . The system of claim 5 , wherein determining if noise is present comprises:
determining whether the signal value is above a signal threshold; and determining whether the noise value is above a noise threshold.

7 . The system of claim 1 wherein the uneven illumination detector module is further configured to:
create a blurred image plane;
determine a mean of the blurred image plane;
determine a standard deviation of the blurred image plane;
subtract the mean of the blurred image plane from the blurred image plane, and configured to add a fraction of that result to the image to create a calculated image plane;
determine the maximum value of the calculated image plane;
determine the minimum value of the calculated image plane;
determine the mean value of the calculated image plane;
determine the standard deviation of the calculated image plane; and
determine whether the image has uneven illumination.

8 . The system of claim 7 , wherein determining whether the stored gray scale image has uneven illumination image comprises:
determining whether the maximum value in the calculated image plane is greater than a threshold value.

9 . The system of claim 1 , wherein the blurriness detector module is configured to:
calculate a threshold percentile histogram value of the stored gray scale image; determine a minimum value of the pixels in the stored gray scale image; determine a maximum value of the pixels in the stored gray scale image; determine a mean value of the stored gray scale image; determine the standard deviation of the stored gray scale image; and determine whether the stored gray scale image is blurry.

10 . The system of claim 10 , wherein determining whether the stored gray scale image is blurry comprises:
evaluating whether the standard deviation is less than or equal to a threshold.

11 . The system of claim 1 , wherein the output indicator is configured to indicate:
a failure state if a blurriness flag is set as true; a caution state if an uneven illumination flag is set as true; a caution state if a hotspot flag is set as true; a caution state if a noise flag is set as true; and a pass state if no flags are set as true.

12 . A system for measuring image quality comprising:
a digital imaging system configured to capture an image; a data storage system configured to store the captured image as a gray scale image; a processing system configured to access the stored gray scale image, wherein the processor system includes a blurriness detection module, the blurriness detection module configured to detect the presence of blurriness in the stored gray scale image; and an output indicator configured to notify a user of a fail condition if blurriness is detected.

13 . The system of claim 12 , further comprising:
a hotspot detector configured to detect the presence of a hotspot in the image; a noise detector configured to detect the presence of noise in the image; and an uneven illumination detector for detecting the presence of uneven illumination in the image.

14 . The system of claim 13 , wherein the output indicator is further configured to indicate:
a caution state if an uneven illumination is detected; a caution state if a hotspot is detected; a caution state if a noise is detected; and a pass state if no blurriness, noise, hotspot, or uneven illumination is detected.

15 . A method for measuring image quality comprising:
capturing a digital image with a digital imaging system; storing the captured image as a gray scale image within a data storage system; accessing the stored gray scale image with a processing system, wherein the processing system:
detects the presence of a hotspot in the image and sets a hotspot flag as true if a hotspot is detected;
detects the presence of noise in the image and sets a noise flag as true if noise is detected;
detects the presence of uneven illumination in the image and sets a uneven illumination flag as true if uneven illumination is detected;
detects the presence of blurriness in the image and sets a blurriness flag as true if blurriness is detected; and
notifying a user with an output indicator if the hotspot, noise, uneven illumination, or blurriness flags were set as true.

16 . The method of claim 15 wherein the image further comprises text.

17 . The method of claim 15 wherein the image further comprises a color image, and wherein the image is converted into a gray scale image with a processing system.

18 . The method of claim 15 wherein detecting a hotspot further comprises:
determining a minimum value of the pixels in the stored gray scale image;
determining a maximum value of the pixels in the stored gray scale image;
determining a mean value for the stored gray scale image;
determining a standard deviation for the stored gray scale image; and
evaluating the maximum value, minimum value, mean value, and standard deviation to determine whether a hotspot is present.

19 . The method of claim 15 wherein detecting the presence of noise further comprises:
applying a median blur to an image plane of the stored gray scale image to create a signal plane;
subtracting a signal plane from the image plane to create a residual noise plane;
determining signal and noise values for the image; and
determin

Assignees

• Kodak Alaris Inc

Inventors

• Beato Louis

Classifications

• G06T7/0002 Primary

  Inspection of images, e.g. flaw detection · CPC title

• G06F18/2413

  based on distances to training or reference patterns · CPC title

• G06F18/24

  Classification techniques · CPC title

• G06T2207/30168

  Image quality inspection · CPC title

• G06T2207/10004

  Still image; Photographic image · CPC title