Method for capturing images comprising a measurement of local motions

• US 20060228049A1
• Filed: 02/14/2006
• Published: 10/12/2006
• Est. Priority Date: 02/17/2005
• Status: Active Grant

• Current Assignee: Stmicroelectronics France (STMicroelectronics NV)
• Original Assignee: STMicroelectronics SA (STMicroelectronics NV)
• Inventors: Martin, Lionel, Gensolen, Fabrice, Cathebras, Guy
• Granted Patent: US 7,925,051 B2
• Time in Patent Office: Days
• US Class Current: 382/309
• CPC Class Codes: G06T 7/20 Analysis of motion motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51
• 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
• CPC Class Codes: H04N 23/6812 based on additional sensors, e.g. acceleration sensors
• CPC Class Codes: H04N 23/682 Vibration or motion blur correction

Abstract

A method for capturing a sequence of video images, by means of an imager including an estimation of the parameters of a model of global motion between successive images. In one embodiment, the method comprises a measurement of local motions on edges of the images, and the estimation of the parameters of the global motion model is performed using the result of the measurement of local motions on the edges of the images.

Claims

1. A method for capturing video images by means of an imager, comprising:

measuring local motions on edges of the images; and

estimating parameters of a global motion model using results of the measurement of local motions on the edges of the images.

2. The method of claim 1, further comprising:

calculating local motions using the estimated parameters of the global motion model;

comparing the local motions calculated and the local motions measured on the edges of the image; and

determining whether the estimated parameters of the global motion model at least partially coincide with the local motions measured.

3. The method of claim 1, wherein the measurement of local motions is taken on four edges of the image.

4. The method of claim 1, wherein the measurement of local motions on the edges of the images is taken according to a single motion detection axis parallel to the edges of the images, and according to two directions of displacement along the detection axis.

5. The method of claim 1, comprising providing and using an imager comprising:

a central region for capturing images comprising light-sensitive cells; and

a peripheral region equipped with motion sensors, supplying local motion data.

6. The method of claim 5, wherein the peripheral region completely surrounds the central region.

7. The method of claim 5, wherein the peripheral region comprises groups of motion sensors each comprising a single row of motion sensors.

8. The method of claim 5, wherein the peripheral region comprises unidirectional motion sensors having a single motion detection axis and two displacement detection directions along the motion detection axis.

9. The method of claim 5, wherein:

the capturing of the images is performed in progressive mode, and

acquisition of local motion data associated with an image is performed in progressive mode and extends in a time window delimited by a first instant corresponding substantially to a start of the capture of a previous image and a second instant corresponding substantially to an end of the capture of the image.

10. The method of claim 1, comprising:

capturing a first and a second image, wherein

measuring local motions comprises measuring local motions at points of a first set of image points, during an acquisition period extending between the capture of the first image and the capture of the second image; and

estimating parameters of a global motion model comprises;

calculating presumed positions of the image points in the first set of image points on the second image according to their positions on the first image, using the local motions measured on the edges of the image, to obtain a second set of image points;

calculating initial parameters of the global motion model using the second set of image points;

calculating theoretical positions of the image points in the first set on the second image according to their positions on the first image, using the initial parameters of the global motion model, to obtain a third set of image points;

comparing the second and third sets of image points; and

removing, from the first set of image points, image points having a deviation greater than a first threshold between their positions in the second set of image points and their positions in the third set of image points, if the number of image points to be removed is above a second threshold.

11. The method of claim 10, comprising calculating new parameters of the global motion model, after removing image points from the first set of image points.

12. The method of claim 1, wherein the parameters of the global motion model comprise an angle of rotation, a zoom factor, and two parameters of translation according to orthogonal axes.

13. The method of claim 1, comprising a step of stabilizing the video images conducted using the estimated parameters of the global motion model.

14. The method of claim 1, further comprising:

compensating for local motions attributable to an animated object entering the images.

15. A video imager comprising:

a first region equipped with light-sensitive cells for capturing images; and

a second region on an edge of the first region, equipped with motion sensors enabling local motions to be measured on the edges of the images simultaneously with the capture of the images, wherein the video imager is configured to supply local motion data in addition to video data proceeding from the light-sensitive cells.

16. The imager according to claim 15, comprising circuits for collecting local motion data supplied by the motion sensors, and for supplying a temporal series of local motion data.

17. The imager according to claim 15, wherein the first region comprises circuits for image acquisition in progressive mode, and the second region comprises circuits for local motion data acquisition in progressive mode, the acquisition of the local motion data associated with a current image extending in a time window delimited by a first instant corresponding substantially to a start of the capture of a previous image and a second instant corresponding substantially to an end of the capture of the current image.

18. The imager according to claim 15, wherein the second region surrounds the first region.

19. The imager according to claim 15, wherein the second region comprises groups of motion sensors each comprising a single row of sensors.

20. The imager according to claim 15, wherein the second region comprises unidirectional motion sensors having one motion detection axis and two motion detection directions along the detection axis.

21. The imager according to claim 15, wherein each motion sensor comprises:

at least two photoreceptors supplying electric signals varying according to a light intensity received,

a correlation circuit for producing at least one local motion signal according to the electric signals, and

an integrator circuit for supplying a mean value of the local motion signal, cyclically receiving a reset signal.

22. A video image acquisition device, comprising:

an imager having;

a first region equipped with light-sensitive cells for capturing images; and

a second region on an edge of the first region, equipped with motion sensors enabling local motions to be measured on the edges of the images simultaneously with the capture of the images, wherein the video imager is configured to supply local motion data in addition to video data; and

a calculating device programmed to estimate parameters of a model of global motion between successive images supplied by the imager, using a result of a measurement of local motions on edges of the images taken by the imager.

23. The device according to claim 22, wherein the calculating device is configured to estimate the parameters of the global motion between a first and a second image by:

from an initial set of image points on the first image, calculating presumed positions of the image points on the second image according to their positions on the first image, using the local motions measured by the imager, to obtain a second set of image points;

calculating initial parameters of the global motion model using the second set of image points;

calculating theoretical positions of the image points in the initial set on the second image according to their positions on the first image, using the initial parameters of the global motion model, to obtain a third set of image points;

comparing the second and third sets of image points; and

selectively removing from the first set of image points, image points having a deviation greater than a first threshold between their positions in the second set and their positions in the third set, when the number of image points having the deviation greater than the first threshold is above a second threshold.

24. The device according to claim 23, wherein the calculating device is configured to calculate new parameters of the global motion model, after image points having the deviation greater than the first threshold have been removed from the first set of image points.

25. The device according to claim 22, wherein the parameters of the global motion model comprise an angle of rotation, a zoom factor, and two parameters of translation according to orthogonal axes.

26. The device according to claim 22, wherein the calculating device is configured to stabilize video images using the estimated parameters of the global motion model.

27. A video imager comprising:

means for capturing a first video image and a second video image; and

means for measuring local motions associated with a portion of the captured video images.

28. The video imager of claim 27, wherein:

the means for capturing the first and second video images comprises a first plurality of light-sensitive cells; and

the means for measuring local motions comprises a second plurality of cells, each cell in the second plurality of cells comprising a plurality of photoreceptors and means for providing an indication of a displacement speed of an image point.

29. The video imager of claim 28, wherein the second plurality of cells surrounds the first plurality of cells.

30. The video imager of claim 28, wherein the means for providing the indication of the displacement speed comprises:

a first edge detector coupled to a first one of the plurality of photoreceptors;

a first pulse shaper coupled to the first edge detector;

a second edge detector coupled to a second one of the plurality of photoreceptors; and

a second pulse shaper coupled to the second edge detector.

31. A computer-readable memory medium containing instructions for causing an imaging device to process video images by generating control signals to cause the imaging device to:

measure local motions on edges of the images; and

estimate parameters of a global motion model using results of the measurement of local motions on the edges of the images.

32. The computer-readable memory medium of claim 31, wherein the instructions further cause the imaging device to:

capture a first and a second image, wherein

measuring local motions comprises measuring local motions at points of a first set of image points during an acquisition period extending between the capture of the first image and the capture of the second image; and

estimating parameters of a global motion model comprises;

calculating a second set of image points using the local motions measured on the edges of the image;

calculating initial parameters of the global motion model using the second set of image points;

calculating a third set of image points using the initial parameters of the global motion model;

comparing the second and third sets of image points; and

modifying the initial parameters of the global motion model based on the comparison of the second and third sets of image points.