Image processor for target detection and tracking

Claim 2

An image processor for producing updated intensity images for use in detecting and tracking a plurality of targets, said image processor comprising: a first and a second plurality of memory locations for storing intensity images; and a signal processor, responsive to a plurality of externally derived intensity images, for producing the updated intensity images, said signal processor being coupled to said first and said second plurality of memory locations, said externally derived intensity images including a plurality of reference images, said first plurality of memory locations for storing said reference images, respective pairs of said reference images including information about the location and the change in location of at least one of said targets during a predetermined time interval, and said externally derived intensity images including at least one intensity image incorporating target detection signals received providing information about the locations of said targets during a particular time interval relative to said predetermined time interval, said second plurality of memory locations for storing said at least one intensity image; said signal processor comprising a closed loop feedback system for providing a plurality of images for updating said reference images in response to said at least one intensity image, said system including: a first filter responsive to said plurality of reference images for producing a simulated image of said at least one intensity image; and an image generator for forming a plurality of images for updating said reference images by combining said simulated image with said at least one intensity image.

Claim 3

An image processor for producing updated intensity images for use in detecting and tracking a plurality of targets, said image processor comprising: a first and second plurality of memory locations for storing intensity images; and a signal processor, responsive to a plurality of externally derived intensity images, for producing the updated intensity images, said signal processor being coupled to said first and said second plurality of memory locations, said externally derived intensity images including a plurality of reference images, said first plurality of memory locations for storing said reference images, respective pairs of said reference images including information about the location and the change in location of at least one of said targets during a predetermined time interval, and said externally derived intensity images including at least one intensity image incorporating target detection signals received providing information about the locations of said targets during a particular time interval relative to said predetermined time interval, said second plurality of memory locations for storing said at least one intensity image; said signal processor comprising a closed loop feedback system for producing a plurality of images for updating said reference images in response to said at least one intensity image, said system including: a first filter responsive to said plurality of reference images for producing a simulated intensity image of said at least one intensity image; an image generator for forming a plurality of combined images by combining said simulated intensity image with said at least one intensity image; and a second filter responsive to said combined images for providing said plurality of images for updating said reference images.

Claim 4

An image processor for producing updated intensity images for use in detecting and tracking a plurality of targets, said image processor comprising: a first and second plurality of memory locations for storing intensity images; and a signal processor, responsive to a plurality of externally derived intensity images, for producing the updated intensity images, said signal processor being coupled to said first and said second plurality of memory locations, said externally derived intensity images including a plurality of reference images, said first plurality of memory locations for storing said reference images, respective pairs of said reference images including information about the location and the change in location of at least one of said targets during a predetermined time interval, and said externally derived intensity images including at least one intensity image incorporating target detection signals received providing information about the locations of said targets during a particular time interval relative to said predetermined time interval, said second plurality of memory locations for storing said at least one intensity image; said signal processor comprising a closed loop feedback system for producing a plurality of images for updating said reference images in response to said at least one intensity image, said system including: a first filter responsive to said plurality of reference images for producing a simulated intensity image of said at least one intensity image; an image generator for forming a plurality of combined images by combining said simulated intensity image with said at least one intensity image; and a second filter responsive to said combined images for providing said plurality of images for updating said reference images.

Claim 5

The image processor of claim 2, wherein said signal processor comprises a digital signal processor, each of said images for updating said reference images, and said reference images, being comprised of pixels, each of said pixels having an intensity value.

Claim 6

The image processor of claim 2, wherein said first plurality of memory locations for storing said reference images as compressed intensity images; said image processor further comprising a signal processor for producing reference images from compressed intensity images and for producing compressed intensity images from reference images.

Claim 7

The image processor of claim 3, wherein said first filter comprises a prediction filter for producing a plurality of predicted images from said reference images and a sensor simulation filter for producing said simulated intensity image from said plurality of predicted images.

Claim 8

The image processor of claim 5, wherein said second filter comprises a prediction filter for producing said plurality of images for updating said reference images from said combined images.

Claim 9

The image processor of claim 6, wherein each of said prediction filters comprises a Kalman prediction filter.

Claim 10

The image processor of claim 6, wherein said image generator for forming said plurality of combined images by separately combining each of said predicted images with said at least one intensity image and said simulated intensity image so that the differences between said at least one intensity image and said simulated image are incorporated into said plurality of combined images.

Claim 11

The image processor of claim 8, wherein said image generator for forming said combined images in accordance with the equation C.sub.k =P.sub.k ×D.sub.k /S.sub.k, where C k ,P k ,D k , and S k are image intensities for corresponding pixels of, respectively, the combined image being formed, one of said predicted images, said at least one intensity image, and said simulated intensity image.

Claim 12

The image processor of claim 8, wherein said digital signal processor comprises a microprocessor.

Claim 15

The method of claim 11, and further comprising, before the producing step, the steps of: receiving target detection signals for said plurality of targets; and converting said received target detection signals into said at least one intensity image.

Claim 16

The method of claim 11, and, following the forming step, further comprising the steps of: producing a plurality of updated reference images from said combined intensity images by filtering said combined intensity images; and recursively performing the previously recited steps of producing a simulated intensity image, forming a plurality of combined intensity images, and producing a plurality of updated reference images until the differences between the last produced simulated intensity image and said converted intensity image converge to a predetermined level.

Claim 17

The method of claim 13, and further comprising the additional steps of: producing a residual intensity image corresponding to the differences between said last produced simulated image and said converted image after the convergence to a predetermined level; forming a pair of reference images from said residual intensity image providing information about location and change in location of another target; and recursively performing the steps of producing another simulated intensity image, forming a plurality of combined intensity images, and producing a plurality of updated reference images until the differences between the last produced another simulated intensity image and said converted intensity image converge to a predetermined level.

Claim 18

The method of claim 13, and further including the step of storing said reference images in said first plurality of memory locations as compressed intensity images; and wherein the step of producing a simulated intensity image further includes the step of forming reference images from said compressed intensity images; and wherein the step of producing a plurality of updated reference images from said combined images further includes the step of forming compressed updated reference images from said updated reference images.

Claim 19

The method of claim 13, wherein at least another of said externally derived images is comprised of converted target detection signals received, said at least another externally derived image constituting another converted intensity image, the last-recited converted target detection signals providing information about locations of said targets during another time interval; wherein the step of producing a simulated intensity image comprises producing a simulated intensity image of said another converted intensity image; wherein the forming step comprises forming a plurality of combined intensity images by combining each of said simulated intensity images with the corresponding converted intensity image; and wherein the step of recursively performing the previously recited steps is performed until the differences between each of said last produced simulated intensity images and the corresponding converted intensity image converge to a predetermined amount.

Claim 20

The method of claim 13, wherein the step of producing a simulated image comprises the steps of: producing a plurality of images predicting the locations of said targets during said particular time interval by applying said reference images to a prediction filter; forming a target prediction image by superpositioning said plurality of predicted images; and applying said target prediction image to a sensor simulation filter in order to produce said simulated image.

Claim 21

The method of claim 17, wherein the step of forming said combined intensity images comprises the step of producing a plurality of combined images in a manner in which each of said predicted images are separately combined with said converted intensity image and said simulated image so that the differences between said converted intensity image and said simulated image are incorporated into said combined intensity images.

Claim 22

The method of claim 18, wherein said combined, predicted, converted and simulated intensity images are comprised of a corresponding number of pixels arranged in a rectangular array, the step of forming said plurality of combined intensity images being in accordance with the equation C.sub.k =P.sub.k ×D.sub.k /S.sub.k where C k ,P k ,D k , and S k are image intensities for corresponding pixels of, respectively, the image being formed by combining intensity images, one of said predicted images, said converted intensity image, and said simulated intensity image.

Claim 23

The method of claim 19, wherein the step of producing said updated reference intensity images comprises the step of applying said combined intensity images to a prediction filter.

Claim 24

The method of claim 20, wherein said target detection signals comprise vibrational energy signals.

Claim 25

The method of claim 21, wherein said vibrational energy signals comprise sonic signals.

Claim 26

The method of claim 21, wherein said vibrational energy signals comprise electromagnetic signals.