IMAGE SYSTEM AND DENOISING METHOD THEREOF

• US 20120212652A1
• Filed: 02/13/2012
• Published: 08/23/2012
• Est. Priority Date: 02/18/2011
• Status: Active Grant

• Current Assignee: PixArt Imaging Incorporated
• Original Assignee: PixArt Imaging Incorporated
• Inventors: HSU, En Feng, CHUNG, Ching Lin
• Granted Patent: US 8,823,839 B2
• Time in Patent Office: Days
• US Class Current: 348/241
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 5/50 using two or more images, e.g. averaging or subtraction
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: H04N 23/745 Detection of flicker frequency or suppression of flicker wherein the flicker is caused by illumination, e.g. due to fluorescent tube illumination or pulsed LED illumination

Abstract

There is provided a denoising method of an image system that detects a brightness variation cycle of ambient light sources, records a brightness variation of at least one noise area within the brightness variation cycle and compensates the brightness variation from current images thereby removing interference from the ambient light sources. There is further provided an image system.

Claims

1. An image system, comprising:

at least one image sensor configured to continuously acquire images; and

a processing unit for determining at least one noise area according to a plurality of images acquired by the image sensor, and recording a brightness variation parameter and a phase parameter of the noise area within a reference brightness variation cycle to accordingly compensate a brightness of the noise area in current images.

2. The image system as claimed in claim 1, wherein the brightness variation parameter is a brightness value of the noise area in each of the images, or a brightness difference or a brightness ratio of the brightness value of the noise area and an average brightness in each of the images acquired by the image sensor within the reference brightness variation cycle;

and the phase parameter is a sampling phase of each of the images acquired by the image sensor within the reference brightness variation cycle.

3. The image system as claimed in claim 1, wherein the processing unit identifies an image area in the images having a fixed brightness variation frequency as the noise area, or identifies an image area in the images having a brightness variation frequency identical to a predetermined brightness variation frequency as the noise area.

4. The image system as claimed in claim 1, wherein the processing unit further records a brightness variation of the noise area for a predetermined time interval so as to identify a brightness variation cycle of ambient light.

5. The image system as claimed in claim 4, wherein after recording the brightness variation of the noise area for the predetermined time interval, the processing unit further identifies whether a brightness of the ambient light varies periodically according to the brightness variation of the noise area recorded within the predetermined time interval.

6. A denoising method of an image system, for removing interference from ambient light having a brightness variation cycle, the denoising method comprising the steps of:

acquiring two images using an image sensor;

calculating a difference of the two images to determine at least one noise area in the images;

recording a brightness variation of the noise area within a reference brightness variation cycle; and

acquiring two current images and calculating a current image difference, retrieving the brightness variation of the noise area of two neighboring images in the reference brightness variation cycle corresponding to sampling phases of the two current images in the brightness variation cycle, and subtracting the retrieved brightness variation from the current image difference.

7. The denoising method as claimed in claim 6, further comprising:

recording a brightness variation of the noise area for a predetermined time interval so as to identify the brightness variation cycle.

8. The denoising method as claimed in claim 7, wherein in the step of recording a brightness variation of the noise area, a brightness value of the noise area in every image or a brightness difference of the noise area between every two images acquired by the image sensor within the predetermined time interval is recorded.

9. The denoising method as claimed in claim 7, wherein the step of recording a brightness variation of the noise area for a predetermined time interval further comprises the steps of:

identifying whether a brightness of the ambient light varies periodically according to the brightness variation of the noise area recorded within the predetermined time interval;

identifying the brightness variation cycle when the brightness of the ambient light varies periodically; and

acquiring a plurality of current images using the image sensor to accordingly determine a system parameter when the brightness of the ambient light does not vary periodically.

10. The denoising method as claimed in claim 6, wherein the step of recording a brightness variation of the noise area within a reference brightness variation cycle is to record a brightness value of the noise area in each of the images or a brightness difference of the noise area between every two neighboring images acquired by the image sensor within the reference brightness variation cycle.

11. The denoising method as claimed in claim 6, wherein the brightness variation of the noise area within the reference brightness variation cycle comprises a reference brightness variation of the noise area between any two neighboring images within the reference brightness variation cycle and a brightness ratio of the brightness variation of the noise area between other two neighboring images with respect to the reference brightness variation.

12. The denoising method as claimed in claim 11, wherein the reference brightness variation is a largest brightness variation of the noise area between two neighboring images within the reference brightness variation cycle.

13. The denoising method as claimed in claim 6, wherein the brightness variation of the noise area within the reference brightness variation cycle comprises an average brightness variation of the noise area between every two neighboring images acquired within the reference variation cycle and a brightness ratio of the brightness variation of the noise area between every two neighboring images with respect to the average brightness variation.

14. A denoising method of an image system, comprising the steps of:

acquiring, using an image sensor, a plurality of images to determine at least one noise area;

recording a brightness variation parameter and a phase parameter of the noise area within a reference brightness variation cycle; and

compensating a brightness of the noise area in current images acquired by the image sensor according to the brightness variation parameter and the phase parameter.

15. The denoising method as claimed in claim 14, further comprising the step of:

recording a brightness variation of the noise area for a predetermined time interval so as to identify a brightness variation cycle of the noise area.

16. The denoising method as claimed in claim 14, wherein the brightness variation parameter is a brightness value of the noise area in each of the images, or a brightness difference or a brightness ratio of the brightness value of the noise area and an average brightness in each of the images acquired by the image sensor within the reference brightness variation cycle;

and the phase parameter is a sampling phase of each of the images acquired by the image sensor within the reference brightness variation cycle.

17. The denoising method as claimed in claim 14, wherein the step of acquiring, using an image sensor, a plurality of images to determine at least one noise area is to identify an image area in the images having a fixed brightness variation frequency as the noise area, or to identify an image area in the images having a brightness variation frequency identical to a predetermined brightness variation frequency as the noise area.

18. The denoising method as claimed in claim 14, wherein the step of acquiring, using an image sensor, a plurality of images to determine at least one noise area further comprises the steps of:

acquiring two images using the image sensor; and

calculating a difference of the two images to determine at least one noise area in the images acquired.

19. The denoising method as claimed in claim 14, wherein the step of compensating a brightness of the noise area in current images acquired by the image sensor according to the brightness variation parameter and the phase parameter is to acquire two current images and calculate a current image difference, to retrieve a brightness variation of the noise area of two neighboring images in the reference brightness variation cycle corresponding to sampling phases of the two current images in the brightness variation cycle, and to subtract the retrieved brightness variation from the current image difference.

20. The denoising method as claimed in claim 14, further comprising:

recording a brightness variation of the noise area for a predetermined time interval so as to identify whether the brightness variation of the noise area varies periodically.