Visualization of defects in a frame of image data

Patent metadata

Field	Value
Publication number	US-9412030-B2
Application number	US-201414564399-A
Country	US
Kind code	B2
Filing date	Dec 9, 2014
Priority date	Dec 9, 2008
Publication date	Aug 9, 2016
Grant date	Aug 9, 2016

Abstract

Disclosed are systems, computer-readable mediums, and methods for receiving, from a camera, a frame representing image data prior to the image data being saved in long term memory. Areas of image data are analyzed to determine types of defects contained within each area of image data. At least one area contains a defect, and the types of defects include glare, blur, defocused, and noise. The frame is visually altered based upon each area of image data that contains a defect such that each defect can be ascertained within the frame, and the altered frame is displayed.

First claim

What is claimed is:

1. A method for visualizing defects in a frame of image data, the method comprising:
receiving, from a camera, a frame representing image data prior to the image data being saved in long term memory;
analyzing, using a processor, areas within image data to determine types of defects contained within each area within the image data, wherein at least one area contains a defect, and wherein the types of defects comprise glare, blur, defocused, and noise;
visually altering the frame based upon each area of image data that contains a defect such that each defect can be ascertained within the frame;
displaying the altered frame;
determining a color for each area of image data based upon a type of defect contained within the area of image data, wherein each type of defect has a distinct color;
coloring, using the color, each area that contains a defect;
determining for each area of image data a level of defect;
determining a color gradient for each area of image data based upon the level of defect; and
using the color gradient to color each area that contains a defect.

2. The method of claim 1 , further comprising segmenting the frame into non-overlapping areas.

3. The method of claim 1 , further comprising coloring, using a unique color, each area that does not contain a defect.

4. The method of claim 1 , further comprising further altering the frame to include a grid that visually represents the areas of image data.

5. The method of claim 4 , wherein the grid comprises hexagons.

6. The method of claim 1 , wherein analyzing each area of image data to determine types of defects contained within the area of image data comprises:
determining if each area of the image data includes a glare defect; determining if each area of the image data includes a blur defect;
determining if each area of the image data includes a defocus defect; and
determining if each area of the image data includes a noise defect.

7. The method of claim 1 , further comprising:
receiving an activation signal to save image data to the long term memory;
determining a level of defects within the frame based upon the analyzing of each area of image data to determine types of defects contained within the area of image data;
determining the level of defects within the frame is below a predetermined threshold; and
saving the frame in the long term memory based upon the determining the level of defects within the frame is below a predetermined threshold.

8. The method of claim 1 , wherein analyzing each area of image data to determine types of defects contained within the area of image data comprises using one or more classifiers, and wherein the method further comprises:
receiving data indicating an additional undetected defect in the image data; and
updating parameters of the one or more classifiers based upon the data indicating the additional undetected defect in the image data, wherein the updated parameters train the one or more of the classifiers to detect the additional undetected defect in the image data.

9. The method of claim 8 , further comprising:
sending the image data to a remote computing device;
receiving optical character recognition data, from the remote computer device, identifying text within the image data, wherein the optical character recognition data comprises a level of incorrectly or uncertainly recognized characters in the image data, and wherein the optical character recognition data is the data indicating the additional undetected defect in the image data;
determining the level of incorrectly or uncertainly recognized characters in the image data is above a predetermined threshold value; and
updating the parameters of the one or more classifiers based upon the determining the level of incorrectly or uncertainly recognized characters in the image data is above the predetermined threshold value.

10. The method of claim 8 , further comprising:
enabling graphical user interface functions related to training the classifiers; and
receiving input from a graphical user interface that changes the area of the detected defect, wherein the received input is the data indicating the additional undetected defect in the image data.

11. The method of claim 10 , wherein the area of the detected defect is enlarged based upon the received input.

12. The method of claim 10 , wherein the area of the detected defect is reduced based upon the received input.

13. The method of claim 12 , wherein the area of the detected defect is reduced to zero.

14. The method of claim 8 , further comprising:
enabling graphical user interface functions related to training the classifiers; and
receiving input from a graphical user interface that changes a type of the detected defect, wherein the received input is the data indicating the additional undetected defect in the image data.

15. A system to visualize defects in a frame of image data, the system comprising:
a camera configured to capture image data;
one or more electronic processors configured to:
receive, from the camera, a frame representing image data prior to the image data being saved in long term memory;
analyze areas of image data to determine types of defects contained within each area of image data, wherein at least one area contains a defect, and wherein the types of defects comprise glare, blur, defocused, and noise;
visually alter the frame based upon each area of image data that contains a defect such that each defect can be ascertained within the frame; and a display configured to display the altered frame;
determine a color for each area of image data based upon a type of defect contained within the area of image data, wherein each type of defect has a distinct color; and
color using the color each area that contains a defect.

16. The system of claim 15 , wherein the one or more processors are further configured to color using a unique color each area that does not contain a defect.

17. The system of claim 15 , wherein the one or more processors are further configured to:
determine for each area of image data a level of defect;
determine a color gradient for each area of image data based upon the level of defect; and

use the color gradient to color each area that contains a defect.

18. The system of claim 15 , wherein to analyze each area of image data to determine types of defects contained within the area of image data the one or more processors are further configured to:
determine if each area of the image data includes a glare defect;
determine if each area of the image data includes a blur defect;
determine if each area of the image data includes a defocus defect; and
determine if each area of the image data includes a noise defect.

19. The system of claim 15 , further comprising:
a long term memory configured to store image data, wherein the one or more processors are further configured to:
receive an activation signal to save image data to the long term memory;
determine a level of defects within the frame based upon the analyzing of each area of image data to determine types of defects contained within the area of image data;
determine the level of defects within the frame is below a predetermined threshold; and
save the frame in the long term memory based upon the determined level of defects within the frame being below a predetermined threshold.

20. A non-transitory computer-readable medium having instructions stored thereon to visualize defects in a frame of image data, the instructions comprisi

Assignees

• Abbyy Dev Llc

Inventors

• Bocharov Konstantin
• Kostyukov Mikhail

Classifications

• G06V30/164

  Noise filtering · CPC title

• G06V30/10

  Character recognition · CPC title

• G06T2207/20056

  Discrete and fast Fourier transform, [DFT, FFT] · CPC title

• G06K2209/01

  Physics · mapped topic

• G06T2207/20201

  Motion blur correction · CPC title