Methods and systems for detecting cracks in illuminated electronic device screens

• US 10,269,110 B2
• Filed: 06/28/2016
• Issued: 04/23/2019
• Est. Priority Date: 06/28/2016
• Status: Active Grant

• Current Assignee: Ecoatm Incorporated (Coinstar Asset Holdings, LLC)
• Original Assignee: Ecoatm Incorporated (Coinstar Asset Holdings, LLC)
• Inventors: Forutanpour, Babak, Ploetner, Jeffrey
• Primary Examiner(s): Bitar, Nancy
• Application Number:
  US15/195,828
• Publication Number:
  US 20170372465A1
• Time in Patent Office: 1,029 Days
• Field of Search: None
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06F 18/24 Classification techniques
• CPC Class Codes: G06T 2207/20016 Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
• CPC Class Codes: G06T 2207/30108 Industrial image inspection
• CPC Class Codes: G06T 2207/30121 CRT, LCD or plasma display
• CPC Class Codes: G06T 7/0004 Industrial image inspection
• CPC Class Codes: G06T 7/001 using an image reference approach
• CPC Class Codes: G06T 7/40 Analysis of texture depth or shape recovery from texture G06T7/529

Abstract

Systems and methods for detecting the cracks in illuminated electronic device screens are disclosed. In one embodiment, the method includes receiving an image of an electronic device screen and retrieving a plurality of kernels, each having values corresponding to a line region and a non-line region, with the orientation of the line region and the non-line region differing for each kernel. At least some of the kernels are applied to the image to obtain, at various locations of the image, values corresponding to the line regions and the non-line regions. Based on the values corresponding to the line regions and the non-line regions, cracks are automatically identified in the electronic device screen.

Claims

1. A method performed by one or more computing devices for detecting cracks in a screen of an electronic device, the method comprising:

receiving an image of an electronic device screen;

overlaying a kernel on the image of the electronic device screen at a selected location, the kernel comprising a matrix having a line region and a non-line region, wherein the line region includes a left-line (LL) region and a right-line region (RL), and wherein the non-line region includes an above-the-line (AL) region and a below-the-line region (BL);

determining values associated with brightness at the line region and the non-line region of the kernel at the selected location; and

based on one or more differences in the values associated with brightness at the line region and the non-line regions, automatically identifying cracks in the electronic device screen, wherein automatically identifying cracks comprises identifying a crack if an average brightness of the LL region and the RL region is greater than an average brightness of the AL region and the BL region by at least a predetermined threshold amount.

2. The method of claim 1, further comprising:

generating an image pyramid with multiple layers of the image, each layer being down sampled to a different degree; and

for each layer, applying at least one kernel to the layer to obtain, at various locations of the layer, values corresponding to the line regions and the non-line regions.

3. The method of claim 1 wherein the kernel is configured to detect cracks in the image that align with the line region of the kernel.

4. The method of claim 1 wherein the at least one kernel includes at least 10 kernels, each kernel having a different orientation of the line region and the non-line region.

5. The method of claim 1 further comprising moving the kernel across various locations on the screen and multiplying underlying pixel values of the image with values of the kernel.

6. The method of claim 1 wherein, if no crack is automatically identified for the kernel at the selected location, the method further includes applying a next kernel at the selected location.

7. The method of claim 1 wherein, if a crack is automatically identified for the kernel at the select location, the method further includes applying the kernel at a next location.

8. The method of claim 1 further comprising determining whether at least one of the following conditions is satisfied:

(a) an average brightness of the AL region and an average brightness of BL region are within a predetermined range of one another;

(b) the average brightness of the AL region or the average brightness of the BL region is below a predetermined threshold level;

(C) the average brightness of the LL region and the RL region is greater than the average brightness of the AL region and the BL region by a predetermined threshold amount;

(d) the average brightness of the LL region and the average brightness of the RL region differ from one another by more than a predetermined threshold amount;

(e) any value in the LL, RL, AL, or BL regions is greater than a predetermined threshold value;

(f) any value in the LL, RL, AL, or BL regions is lower than a predetermined threshold value; and

(g) any three consecutive LL or RL values have corresponding brightness levels that do not increase or decrease monotonically,

wherein, if any of conditions (a)-(g) are satisfied, then the identified crack is discarded.

9. The method of claim 1, further comprising, before overlaying the kernel on the image, evaluating the image of the electronic device screen to determine if an animated background is present.

10. A method performed by one or more computing devices for detecting cracks in an illuminated electronic device screen, the method comprising:

receiving an image of an illuminated electronic device screen;

applying a first kernel to a first location of the image and automatically determining whether there is a crack in the first location, wherein the first kernel comprises a matrix having a line region and a non-line region;

if no crack is identified in the first location, then applying a second kernel to the first location of the image and automatically determining whether there is a crack in the first location, wherein the second kernel comprises a matrix having a line region and a non-line region having a different orientation than the first kernel; and

if there is an identified crack in the first location, then applying the first kernel to a second location of the image and automatically determining whether there is a crack in the second location, wherein the line region of each kernel comprises a left-line (LL) region and a right-line region (RL) and the non-line region of each kernel includes an above-the-line (AL) region and a below-the-line region (BL), and wherein automatically determining whether there is a crack comprises identifying a crack if an average brightness of the LL region and the RL region is greater than an average brightness of the AL region and the BL region by at least a predetermined threshold amount.

11. The method of claim 10, further comprising:

generating an image pyramid with multiple layers of the image, each layer being down sampled to a different degree; and

for each layer, applying the first kernel to locations of the layer to automatically determine whether there is a crack at the locations.

12. The method of claim 10 wherein the method comprises applying at least 10 kernels in sequence to a location of the image until a crack is identified at the location.

13. The method of claim 10 wherein, automatically determining whether there is a crack further comprises performing one or more secondary analyses based on the values in one or more of the LL, RL, AL, and BL regions.

14. The method of claim 10, further comprising applying the first kernel to each location of the image and automatically determining whether there is a crack in each of the locations.

15. A computer-readable memory carrying computer-executable instructions for causing one or more processors to facilitate detecting cracks in an illuminated electronic device screen, the computer-executable instructions comprising instructions that, when executed by the one or more processors:

receive an image of an electronic device screen;

overlaying a kernel on the image of the electronic device screen, the kernel comprising a matrix having a line region and a non-line region, wherein the line region includes a left-line (LL) region and a right-line region (RL), and wherein the non-line region includes an above-the-line (AL) region and a below-the-line region (BL);

determining values associated with brightness at the line region and the non-line region of the kernel; and

based on one or more differences in the values associated with brightness at the line region and the non-line regions, automatically identify cracks in the electronic device screen, wherein automatically identifying cracks comprises identifying a crack if an average brightness of the LL region and the RL region is greater than an average brightness of the AL region and the BL region by at least a predetermined threshold amount.

16. The computer-readable memory of claim 15 wherein the instructions, when executed by the one or more processors, further:

generate an image pyramid with multiple layers of the image, each layer being down sampled to a different degree; and

for each layer, apply the kernel with the layer to obtain, at various locations of the layer, values corresponding to the line regions and the non-line regions.

17. The computer-readable memory of claim 15, further comprising moving the kernel to various locations on the screen and multiplying underlying pixel values of the image with values of the kernel.