Blur detection system for night scene images

• US 9,412,039 B2
• Filed: 11/03/2010
• Issued: 08/09/2016
• Est. Priority Date: 11/03/2010
• Status: Active Grant

• Current Assignee: Nikon Corporation
• Original Assignee: Nikon Corporation
• Inventors: Hong, Li, Homma, Itaru
• Primary Examiner(s): Harandi, Siamak
• Application Number:
  US13/823,015
• Publication Number:
  US 20130169824A1
• Time in Patent Office: 2,106 Days
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/30168 Image quality inspection
• CPC Class Codes: G06T 7/0002 Inspection of images, e.g. flaw detection
• CPC Class Codes: G06V 10/993 Evaluation of the quality of the acquired pattern
• CPC Class Codes: H04N 23/6811 based on the image signal
• CPC Class Codes: H04N 23/80 Camera processing pipelines; Components thereof
• CPC Class Codes: H04N 23/81 for suppressing or minimising disturbance in the image signal generation

Abstract

A system (300) for detecting whether an image (16) is blurred performs the steps of: identifying valid light trails in the test image (308), and classifying the test image as blurred if there is a sufficient number of matching valid light trails oriented in a repetitive pattern (310). The system (300) is designed to automatically detect whether a night scene image (16) is blurred based on the existence and characteristics of any light trails (552) in the captured image (16). Because the method extracts the characteristics of the light trails (552) and analyzes these light trails (552), it accurately detects blur in the challenging night scene images and has a relatively low false positive rate.

Claims

1. A method for determining if a test image is blurred, the method comprising the steps of;

reviewing the pixel values in the test image with a control system that includes a processor to identify potential light trail pixels;

reviewing the potential light trail pixels with the control system to identify valid light trails in the test image; and

reviewing the valid light trails with the control system to identify the number of valid light trails with similar orientation and size to estimate if the test image is blurred with the control system.

2. The method of claim 1 further comprising the step of identifying if the test image is a potential night scene prior to the step of reviewing the potential light trail pixels.

3. The method of claim 1 further comprising the step of down sampling the test image prior to the step of reviewing the potential light trail pixels.

4. The method of claim 1, wherein the step of reviewing the potential light trail pixels includes the step of identifying saturated pixels in the image with the control system by identifying pixels having a value that is greater than a predetermined light saturation threshold.

5. The method of claim 4, wherein the step of reviewing the potential light trail pixels includes the step of building potential light trails from connected saturated pixels.

6. The method of claim 5, wherein the step of reviewing the potential light trail pixels includes the step of computing one or more region properties of the potential light trails and identifying valid light trails with the control system based on the region properties of the potential light trails.

7. The method of claim 1, further comprising the step of comparing the size and orientation of the valid light trails.

8. The method of claim 1, further comprising the step of comparing the proximity of the valid light trails.

9. A camera including a capturing system including an image sensor for capturing a test image, and a control system including a processor that utilizes the method of claim 1 to estimate if the test image is blurred.

10. An image apparatus for capturing an image, the image apparatus comprising:

a capturing system including an image sensor for capturing the image; and

a control system including a processor that estimates if the image is blurred by (i) reviewing the pixel values of the pixels in the image to identify potential light trail pixels;

(ii) reviewing the potential light trail pixels to identify valid light trails in the test image, and (iii) reviewing the valid light trails to identify the number of valid light trails oriented in a similar pattern to estimate if the test image is blurred.

11. The image apparatus of claim 10 wherein the control system identifies if the test image is a potential night scene prior to identifying valid light trails.

12. The image apparatus of claim 10 wherein the control system down samples the test image prior to identifying valid light trails.

13. The image apparatus of claim 10 wherein the control system identifies saturated pixels in the image by identifying pixels that have a value that is greater than a predetermined light saturation threshold.

14. The image apparatus claim 13, wherein the control system builds potential light trails from connected saturated pixels.

15. The image apparatus of claim 14, wherein the control system computes one or more region properties of the potential light trails and identifies valid light trails based on the region properties of the potential light trails.

16. The image apparatus of claim 10, wherein the control system compares the size and orientation of the valid light trails.

17. The image apparatus of claim 10, wherein the control system compares the proximity of the valid light trails.

18. A method for evaluating a test image, the method comprising the steps of;

capturing the test image with an image sensor of a camera;

identifying if the test image is a potential night scene with a control system of the camera that includes a processor; and

in the event the test image is identified as a potential night scene, (i) reviewing the pixel values of the pixels of the test image with the control system to identify pixels in the test image having a value that is greater than a predetermined light saturation threshold to identify saturated pixels with the control system, and (ii) reviewing the saturated pixels to identify valid light trails from connected saturated pixels with the control system; and

(iii) comparing the valid light trails with the control system to estimate if the test image is blurred.

19. The method of claim 18 further comprising the step of down sampling the test image prior to the step of reviewing the saturated pixels.

20. The method of claim 18, wherein the step of reviewing the saturated pixels includes the step of computing one or more region properties of the potential light trails with the control system and identifying valid light trails with the control system based on the region properties of the potential light trails.