Image anti-shake in digital cameras

• US 20070177021A1
• Filed: 03/31/2006
• Published: 08/02/2007
• Est. Priority Date: 01/30/2006
• Status: Active Grant

• Current Assignee: Omnivision Technologies Incorporated (Will Semiconductor Co., Ltd. Shanghai)
• Original Assignee: Omnivision Technologies Incorporated (Will Semiconductor Co., Ltd. Shanghai)
• Inventors: Shan, Jizhang, Pan, Hui, Liu, Guansong, Lee, Jess, Flamm, Daniel
• Granted Patent: US 7,593,040 B2
• Time in Patent Office: Days
• US Class Current: 348/208.120
• CPC Class Codes: H04N 23/68 for stable pick-up of the scene, e.g. compensating for camera body vibrations
• CPC Class Codes: H04N 23/6811 based on the image signal

Abstract

A method and system for minimizing the effects of image motion in still cameras is provided. In one embodiment, a method of capturing an image in a digital camera is presented. The method includes calculating a sharpness value related to an image input. The method also includes evaluating the sharpness value to determine image motion. The method further includes capturing a next image input data from the image input responsive to evaluating the sharpness value.

Claims

1. A method of capturing an image in a digital camera, comprising:

receiving a previous image input of a previous image frame and forming a sharpness value of the previous image input;

receiving a current image input of a current image frame and forming a sharpness value of the current image input;

predicting that a next image frame motion will be less than a current image frame motion, the prediction being based at least on the sharpness value of the previous image input, and the sharpness value of the current image input;

deciding to capture a next image frame responsive to the prediction; and

capturing the next image frame.

2. The method of claim 1 further comprising receiving a capture command, wherein the receiving of the previous image input of a previous image frame is after receiving the capture command.

3. The method of claim 1 further comprising receiving a first image input of a first image frame and forming a sharpness value of the first image input, wherein the first image frame is different from the previous image frame, and the receiving of the first image input is before the receiving of the previous image input.

4. The method of claim 3 further comprising receiving a capture command, wherein the receiving of the first image input is before the receiving of the capture command and the receiving of the previous image input is after the receiving of the capture command.

5. The method of claim 3 wherein the first image input is a subframe of the first image frame, the previous image input is a subframe of the previous image frame, and the current image input is a subframe of the current image frame.

6. The method of claim 3 wherein forming a sharpness value comprises a high frequency extracting filter.

7. The method of claim 3 wherein the high frequency extracting filter is a high pass filter.

8. The method of claim 1 wherein forming a sharpness value of an image input selected from the previous image input and the current image input comprises a sum of convolving a high pass filter matrix with the image input.

9. The method of claim 8 wherein the high pass filter matrix has a symmetry.

10. The method of claim 1, wherein deciding to capture the next image frame depends on having a frame sharpness of the current image frame that is greater than the frame sharpness of the previous image frame.

11. The method of claim 1, wherein deciding to capture the next image frame depends on having a frame sharpness of the current image frame that is greater than the frame sharpness of any image frame after the capture command.

12. The method of claim 1 wherein the current image input is a subframe of the current image frame, the previous image input is a corresponding subframe of the previous image frame, and the predicting further depends on a difference between a sharpness value of the subframe of the current image frame and a sharpness value of the corresponding subframe of the previous image frame.

13. The method of claim 12 wherein the previous image frame is an image frame immediately preceding the current image frame.

14. The method of claim 13, wherein the subframe of the current image frame is a last subframe of the current image frame and an absolute value of the difference is less than a selected factor multiplied by a maximum of absolute values of other consecutive sharpness value differences.

15. The method of claim 14 wherein the selected factor depends on the sign of the difference and the selected factor is less than 1.

16. A means for capturing a relatively stable picture in a digital camera, including steps for predicting when the image motion of a next image frame will be less than the image motion of a current image frame, and means for capturing the next image frame based on the prediction.

17. A method of capturing a relatively stable picture in a digital camera, comprising:

calculating a first sharpness value related to an image input;

calculating a second sharpness value related to another image input,

calculating a sharpness value difference comprising the first sharpness value and the second sharpness value,

estimating quality of a next image based on the sharpness value difference; and

capturing a next image from the image input responsive to the estimating of the quality of the next image.

18. The method of claim 17 wherein the first sharpness value comprises the sharpness of a subframe and the second sharpness value comprises the sharpness of another subframe.

19. An apparatus for capturing an image at a time of low motion, comprising:

means for calculating at least two sharpness value differences related to an image input;

means for estimating next image quality based on the sharpness value differences; and

means for capturing a next image from the image input responsive to the means for estimating the image quality.

20. A digital camera for capturing a relatively stable picture, comprising:

a digital image sensor operable to output data associated with an image;

machine readable media operable to store data and instructions;

at least one processor operable to perform the instructions and operate on the data;

the instructions being operable to implement;

a sharpness detector coupled to the image sensor, the sharpness detector operable to evaluate the sharpness of an image;

a predictor, operable to predict image motion of a next image based on the sharpness,

a decision maker, operable to decide to capture a next image and to capture the next image into the machine readable media.

21. The digital camera of claim 20 wherein the machine readable media comprises a line buffer for a relatively small number of lines.

22. The digital camera of claim 20 wherein the sharpness detector comprises convolution of a high pass filter matrix with the data associated with an image.

23. A method of capturing an image in a digital camera, comprising:

partitioning an image frame from a digital image sensor into at least four subframes, the subframes having equal amounts of image data;

receiving a capture command;

receiving current subframe image data from the digital image sensor;

storing a selected number of lines of the current subframe image data in a line buffer;

performing a high frequency filtering comprising convolution of a filter matrix with selected columns of the current subframe image data in the line buffer;

evaluating a sharpness of the current subframe based on the high frequency filtering;

evaluating a current to previous sharpness value difference;

determining a maximum current to previous sharpness value difference;

evaluating a sharpness of an image frame comprising the current subframe image data;

predicting the quality of a next image frame based on the current to previous sharpness value difference and the maximum current to previous sharpness value difference; and

deciding to capture a next image frame depending on the predicting and the sharpness of the image frame comprising the current subframe image data.

24. A device for image capture performing the method of claim 23.