Method and apparatus for detecting the presence of one or more images of a known predetermined kind of scene

• US 7,298,908 B2
• Filed: 12/09/2002
• Issued: 11/20/2007
• Est. Priority Date: 09/24/1998
• Status: Expired due to Fees

• Current Assignee: Secretary of State For Defence
• Original Assignee: Qinetiq Limited (QinetiQ Group PLC)
• Inventors: Lewis, Meirion F., Lowans, Brian S., Wilson, Rebecca A.
• Primary Examiner(s): Johns, Andrew W.
• Assistant Examiner(s): CONOVER, DAMON M
• Application Number:
  US10/314,415
• Publication Number:
  US 20040037462A1
• Time in Patent Office: 1,807 Days
• Field of Search: 382/211, 382/222, 382/210
• US Class Current: 382/211
• CPC Class Codes: G06E 1/02 operating upon the order or content of the data handled
• CPC Class Codes: G06V 10/88 Image or video recognition using optical means, e.g. reference filters, holographic masks, frequency domain filters or spatial domain filters

Abstract

An apparatus for detecting the presence of one or more images of a known kind in a scene is disclosed which comprises a digital input means and an optical output means. The digital input means comprises a capture means which passes a capture image to a first electronic processing means to produce a scene pattern corresponding to a Fourier Transform of the scene image. This pattern is then digitally combined with one or more reference patterns corresponding to Fourier Transforms of a reference image. The resulting combined patterns are then used to modulate a beam of light which may be focused to perform an inverse Fourier Transform providing correlation information. In a preferred embodiment, the combined pattern is displayed on a spatial light modulator. A “smart” spatial light modulator is also disclosed which comprises an array of light modulating devices, each device comprising a detector and at least one pixel, whereby light incident upon the detector produces an output signal which is adapted to control the state of the modulating medium. By suitable processing, the “smart” SLM may be adapted to either attenuate light below a certain intensity or only light above a certain intensity to act as a saturable absorber.

Claims

1. An apparatus for detecting the presence of one or more images of a known predetermined kind in a scene, comprising:

means for producing a scene image corresponding to a scene;

a digital input means comprising;

first electronic processing means for processing at least a part of the scene image to produce a scene pattern corresponding to the binarised Fourier Transform of said at least part of the scene image;

storage means for storing one or more reference patterns, each reference pattern comprising the binarised Fourier Transform of a respective reference image;

second processing means for electronically combining the scene pattern with a reference pattern to produce a combined pattern; and

an optical output means for producing an optical output that is derived from the combined pattern wherein the second processing means comprises a logic unit having an exclusive-or function.

2. An apparatus according to claim 1, in which the optical output means comprises a spatial light modulator (SLM) adapted to produce a mask dependent upon the combined pattern, and a light source which is adapted to produce a beam of light that is modulated by the spatial light modulator (SLM).

3. Apparatus according to claim 2, in which the spatial light modulator (SLM) comprises a high speed light modulating ferro-electric liquid crystal material.

4. Apparatus according to claim 3, wherein, after the SLM has been set to display a list of combined patterns (or more than one list), the SLM is driven with an inverse of the integration of the combined patterns of the list displayed so as to reduce the net bias voltage of each pixel to zero.

5. Apparatus according to claim 2, wherein the SLM comprises an array of pixels or elements which can be switched between a first state and a second state, and in which light is modulated differently by pixels that are in the first state in comparison with those that are in the second state.

6. Apparatus according to claim 5, wherein light emerging from pixels in the first and second states have a phase difference of about 180°

.

7. Apparatus according to claim 2, wherein the SLM is adapted to digitally process the combined pattern used to drive the SLM with a chirp function to simulate a zone plate lens incorporated with the combined pattern.

8. Apparatus according to claim 2, wherein the SLM comprises a smart spatial light modulator comprising an array of light modulating devices, each device comprising a light detector and at least one pixel of a modulating medium, in which the light detector is adapted to measure the intensity of light incident thereupon to produce a light-level output signal, which output signal is adapted to control the modulating medium so as to either (a) suppress low level light, or (b) moderate high level light.

9. Apparatus according to claim 8, wherein the smart SLM is adapted to threshold light so as to pass substantially only light above a predetermined level, and in which a detector receives light via the smart SLM.

10. Apparatus according to claim 8, wherein a CCD camera (or other pixelated detector) is provided behind the one or more pixels of the smart SLM so that each pixel of the camera is aligned with a corresponding pixel of the smart SLM.

11. An apparatus according to claim 1, further including an optical to electronic conversion means which is adapted to produce an electrical output signal dependent upon the pattern of the modulated optical output.

12. Apparatus according to claim 1, wherein the scene pattern is combined with each reference pattern, or with each reference pattern of a subset of reference patterns.

13. Apparatus according to claim 1, wherein the means adapted to produce the scene image comprises a camera.

14. Apparatus according to claim 13 in which selection means is provided for selecting the whole or a chosen part of the camera output to form the scene images.

15. Apparatus according to claim 1, wherein the optical output means includes an optical to electronic conversion means located in the plane where a correlation peak is formed.

16. Apparatus according to claim 15, wherein the optical to electronic conversion means is adapted to produce an electrical output signal from the optical pattern formed by the modulated light in a focal plane.

17. Apparatus according to claim 1, wherein the optical output means includes a Difference of Gaussian optical spatial filter.

18. Apparatus according to claim 1, wherein said scene imaging means is physically split into at least two portions, the first portion comprising a scene capture device which is connected by optical and/or electrical means to a second portion comprising a base unit which is adapted to perform the Fourier Transforms.

19. Apparatus according to claim 18, wherein the second portion is divided into an optical unit and a digital unit interconnected.

20. An apparatus for detecting the presence of one or more images of a known predetermined kind in a scene, comprising:

means for producing a scene image corresponding to a scene;

a digital input means comprising;

first electronic processing means for processing at least a part of the scene image to produce a scene pattern corresponding to the Fourier Transform of said at least part of the scene image,

storage means for storing one or more reference patterns, each reference pattern comprising the Fourier Transform of a respective reference image,

second processing means electronically combining the scene pattern with a reference pattern to produce a combined pattern; and

an optical output means for producing an optical output that is derived from the combined pattern, wherein the output optical means comprises a spatial light modulator (SLM), at least two light emitting devices, a lens, a light emitting device controller and an output detector array wherein each light emitting device is spaced at a different position relative to the optical axis of the lens and the controller is adapted to illuminate each device sequentially and, wherein the light emitting means, SLM and output detector array are positioned such that the output detector array receives light from the light emitting devices which has been modulated by the SLM.

21. Apparatus according to claim 20 in which the detector array comprises one line of detectors.

22. Apparatus according to claim 20 in which said light emitting devices are arranged as a linear array which is substantially orthogonal to the output detector array.

23. Apparatus according to claim 20 wherein said light emitting devices comprise a VCSEL array.

24. An apparatus aligning an alignable member with a desired scene object comprising:

means for producing a scene image corresponding to a scene;

a digital input means comprising;

a scene imaging device for producing a scene pattern corresponding to the binarised Fourier Transform of said at least part of the scene image,

storage means for storing one or more reference patterns, each reference pattern comprising a binarised Fourier Transform of a respective reference image,

a pattern combiner for combining the scene pattern with a reference pattern to produce a binarised combined pattern; and

an inverse Fourier Transform means acting on the binarised combined pattern for producing a correlation output having at least two correlation peaks, and a controller for controlling the alignment of an alignable member with the scene object so as to bring the correlation peaks towards each other, preferably making them substantially coincident.

25. A method of detecting the presence of one or more reference images in a scene comprising the steps of:

producing an image scene corresponding to at least part of a scene;

processing at least a part of the captured image scene to produce a scene pattern corresponding to the binarised Fourier Transform of the captured scene image; and

electronically combining the scene pattern with at least one reference pattern corresponding to the binarised Fourier Transform of a reference object to produce a binarised combined pattern, the combination being performed by an exclusive- or logic operation and producing an optical output derived by the combined pattern.

26. A method according to claim 25, including the further step of applying a chirp encoded pattern to the signal representative of either (a) the combined pattern passed to the SLM to produce the mask or (b) to the signal representative of the reference pattern, or (c) the scene pattern, or (d) to a combination of (a),(b),or (c) so as to focus different correlation peaks in different planes and monitoring the pattern formed in a plane corresponding to just one of the correlation peaks.

27. A method according to claim 26, wherein the chirp pattern is calculated and added (or applied) to each reference, or scene pattern prior to binarisation.

28. A method according to claim 26, wherein the chirp encoded pattern is calculated as a binary pattern and combined with at least one of (a) a binarised reference pattern;

or (b) a binarised scene pattern;

or (c) a binarised combined pattern, or (d) a combination of (a) to (c).

29. A method according to claim 26, wherein a different chirp is encoded on each reference pattern within a reference set.

30. An optical pattern recognition and/or location apparatus comprising:

a combining means for combining a scene pattern representative of the Fourier Transform of a scene image, a reference pattern representative of the Fourier Transform of a reference image and a chirp signal to produce a chirped combined pattern;

a spatial light modulator (SLM) for displaying the chirped combined pattern;

a light source for illuminating the SLM;

a lens acting on light modulated by the SLM; and

a monitoring means for receiving light modulated by the SLM and for detecting the presence and/or location of the reference images in the scene image.

31. An apparatus as claimed in claim 30 wherein the combining means is adapted to combine a chirp signal with one or both of the reference and scene pattern.

32. An apparatus as claimed in claim 30 wherein the combining means is adapted to combine the scene and reference patterns to form a combined pattern and combine the combined pattern with the chirp signal.