Target acquisition and tracking system

• US 5,341,142 A
• Filed: 08/28/1990
• Issued: 08/23/1994
• Est. Priority Date: 07/24/1987
• Status: Expired due to Term

• Current Assignee: Northrop Grumman Systems Corporation (Northrop Grumman Corporation)
• Original Assignee: Northrop Grumman Corporation
• Inventors: Reis, James J., Lucero, Antonio B., Garber, David D., Luk, Anthony L.
• Primary Examiner(s): Gregory, Bernarr E.
• Application Number:
  US07/576,121
• Time in Patent Office: 1,456 Days
• Field of Search: 356/5, 356/28, 356/28.5, 342/52-57, 342/59, 342/61-68, 102/393, 244/3.15-3.19
• US Class Current: 342/64
• CPC Class Codes: F41G 7/2226 comparing the observed data with stored target data, e.g. target configuration data
• CPC Class Codes: F41G 7/2253 Passive homing systems, i.e. comprising a receiver and do not requiring an active illumination of the target
• CPC Class Codes: F41G 7/2293 using electromagnetic waves other than radio waves
• CPC Class Codes: G01S 3/7864 T.V. type tracking systems

Abstract

An automatic target acquisition and tracking system has been developed for a focal plane array seeker. The automatic target acquisition is achieved by three independent target acquisition algorithms, viz., the maximum likelihood classification, the video spatial clustering, and the target-to-interference ratio. Each algorithm operates asynchronously and provides independent target detection results. Target information is then combined hierarchically in a probabilistic fashion and prioritized. The highest priority target is handed off to a dual mode tracker consisting of a minimum absolute difference correlation tracker and a centroid tracker. The dual mode tracker subsequently provides a feedback signal to a proportional navigation system or other guidance/control system for directing the flight path of a munition.

Claims

1. Target detection apparatus comprising:

MLC means for determining joint edge feature data and for mapping the same into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from a training set of known examples of known imagery,

GTIR means for determining target-to-interference ratio matched filter feature data in such a manner so to be independent or anticorrelated to said MLC means, and for mapping the same into a probability of target with confidence bounds based a maximum likelihood statistic experimentally derived from said training set,

VSC means for determining grey level feature data in such a manner as to be independent or anticorrelated singly and jointly to said MLC and GTIR means, and for mapping said grey level feature data into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from said training set, and

HPL means for combining or mapping the target detection outputs from said MLC, GTIR, and VSC means into a composite probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from said training set to select a target.

2. A target detection apparatus as defined in claim 1 further including

means for tracking said selected target.

3. A target detection apparatus as defined in claim 2, wherein said tracking means includes

a MAD tracker and a centroid tracker,

said MAD tracker comprising

means for determining the minimum absolute difference correlation between input data and prior input data, and

said centroid tracker comprising means for optimal thresholding input data and calculating the centroid of the resulting image.

4. A target detection system as defined in claim 3, wherein one of said trackers reacquires if said other tracker loses said target.

5. A target detection apparatus as defined in claim 1, wherein said GTIR comprises

means for determining the average local contrast difference between target related picture elements and background related picture elements,

means for determining the local grey level variation, and

means for generating a target to interference related function from said local contrast difference and said local grey level variations.

6. A target detection apparatus as defined in claim 5, comprising

table look-up means for generating a probability of target function and a confidence function from said target to interference function and said local contrast difference,

said table look-up means comprises a three feature Parzen table array,

means for generating a feature vector input to said three feature Parzen table from the cluster size, maximum target to interference function level and contrast difference at the target cluster center,

means for developing said target cluster size and maximum target to interference function from said target to interference related function, and

means for creating said contrast difference at target cluster center from said target to interference related function and said average contrast difference.

7. A target detection apparatus as defined in claim 6, wherein said means for determining the local grey level variation comprises:

means for averaging the vertical picture element values,

means for averaging the horizontal picture element values, and

means for determining a grey level variation from said vertical picture element value and said horizontal picture element value.

8. A target detection apparatus as defined in claim 6, wherein said means for determining the local grey level variation comprises:

means for averaging the vertical picture element values,

means for averaging the horizontal picture element values, and

means for determining a grey level variation from said vertical picture element value and said horizontal picture element value.

9. A target detection apparatus as defined in claim 1, wherein said MLC means comprises

means for developing an adaptive threshold level for target windows, and

means for producing an edge direction map for each picture element in a target sized window.

10. A target detection apparatus as defined in claim 9, wherein said means for developing an adaptive threshold level for target windows comprises

means for determining the intensity gradient for non-overlapping neighborhoods having less than target size areas,

means for determining the average intensity gradient for the entire image area,

look-up table means for developing an average global gradient from said average entire image area gradient,

means for averaging the intensity gradient of a group of said non-overlapping neighborhoods and,

means for estimating the standard deviation of said group of said non-overlapping neighborhoods,

means for summing a function of said standard deviation of said group of said non-overlapping neighborhoods with said average intensity gradient as a group of said non-overlapping neighborhoods,

look-up table means for developing a function of said standard deviation of said group of said non-overlapping neighborhoods, and

means for producting said average global gradient, said average intensity gradient of said group of said non-overlapping neighborhoods, and said look-up table developed function of said standard deviation of said group of said non-overlapping neighborhoods.

11. A target detection apparatus as defined in claim 9 wherein said means for producing an edge direction map for each picture element in a target sized window comprises:

convolutional means for generating a joint edge map from said edge directional map data,

means for computing histograms for target sized window increments of data for those target sized windows containing data above said adaptive threshold level from said joint edge map data and said gradient intensity data,

said histogram including means for extracting lower, middle, and upper classes of feature vectors from data representing each selected target window, said each class of feature vectors bounded by upper and lower values and having overlapping ranges,

lower, middle, and upper look-up table means for mapping said feature vectors as probability of target functions, and

means for producing a target affirmative indication when the probability is 0.5 or greater and a non-target indication when the probability is less than 0.5.

12. A target detection apparatus as defined in claim 1, wherein said VSC means comprises:

means for developing a grey level histogram of a target sized window,

means for calculating upper and lower threshold levels for said histogram as a function of histogram characteristics,

means for eliminating picture elements falling below the low level value,

means for calculating the centroid of a cluster of pixels exceeding said threshold value,

means for creating a global histogram from the entire video image,

means for calculating a first optimal global histogram from said global histogram and said upper grey level threshold,

means for calculating a second global histogram function from said lower grey level threshold,

means for calculating a target sized histogram from said target sized histogram data and said low level threshold value, and

look-up table means responsive to said upper threshold level global histogram said lower threshold level global histogram said target sized window histogram and said centroid for producing a probability of target function and a confidence function.

13. A target detection apparatus as defined in claim 1, wherein said HPL means comprises

means for determining if an input data cluster contains a probability of target assessment,

means for determining the types of probability assessments,

means for forwarding MLC probability of target data to a one dimensional decision Parzen table,

means for forwarding VSC data to a one dimensional decision Parzen table,

means for merging MLC probability target and VSC probability of target and data,

a two dimensional Parzen table responsive to said merged MLC and VSC data,

means for merging MLC probability of target and GTIR probability of target and data,

a two feature Parzen table responsive to said merged MLC and GTIR data,

a single feature Parzen table responsive to GTIR probability of target data,

means to merge GTIR probability of target and VSC probability of target and data,

a two feature Parzen table responsive to said merged GTIR and VSC probability of target data,

means to merge MLC, GTIR, and VSC probability of target, and

a three feature Parzen table responsive to said merged MLC, GTIR, and VSC probability of target data.

14. Apparatus as defined in claim 13, comprising:

final target selection means for ranking the outputs of said decision Parzen tables representing the entire field of view,

a dual mode tracker responsive to the outputs of said final target selection means for selecting a desired target,

said dual mode tracker including a minimum absolute difference correlation tracker 27 and a centroid tracker,

sensor formatter for digitizing the field of view containing the selected target and its immediate surroundings as a target template,

means for generating a minimum absolute difference function for the field of view represented by said template,

means for compensating for tracking vehicle motion,

means for calculating the correlation between a sensed image and a reference provided by said template, and

means for providing feedback signals to the vehicle direction control system.

15. Target detection method comprising the steps of:

determining joint edge feature data and for mapping the same into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from a training set of known examples of known imagery,

determining target-to-interference ratio matched filter feature data in such a manner so to be independent or anticorrelated to said MLC means, and for mapping the same into a probability of target with confidence bounds based a maximum likelihood statistic experimentally derived from said training set,

determining grey level feature data in such a manner as to be independent or anticorrelated singly and jointly to said MLC and GTIR steps, and for mapping said grey level feature data into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from said training set, and

combining or mapping the target detection outputs from said MLC, GTIR, and VSC steps into a composite probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from said training set.

16. A target detection method as defined in claim 15 further including the step of

tracking said selected target.

17. A target detection method as defined in claim 16, wherein said tracking step includes

determining the minimum absolute difference correlation between input data and prior input data, and

optimal thresholding input data and calculating the centroid of the resulting image.

18. A target detection method as defined in claim 17, wherein one of said trackers reacquires said target if said other tracker loses said target.

19. A method as defined in claim 16, comprising

a final target selection step for ranking the outputs of said decision Parzen tables representing the entire field of view,

providing a dual mode tracker responsive to the outputs of said final target selection means for selecting a desired target,

said dual mode tracker including a minimum absolute difference correlation tracker 27 and a centroid tracker,

a sensor formatter step for digitizing the field of view containing the selected target and its immediate surroundings as a target template,

generating a minimum absolute difference function for the field of view represented by said template,

compensating for tracking vehicle motion,

calculating the correlation between a sensed image and a reference provided by said template, and

providing feedback signals to the vehicle direction control system.

20. A target detection method as defined in claim 15, wherein said GTIR comprises

means for determining the average local contrast difference between target related picture elements and background related picture elements,

means for determining the local grey level variation, and

means for generating a target to interference related function from said local contrast difference and said local grey level variations.

21. A target detection method as defined in claim 20 comprising

generating a probability of target function and a confidence function from said target to interference function and said local contrast difference and storing the same in a look up-table comprising a three feature Parzen table array,

generating a feature vector input to said three feature Parzen table array from the cluster size maximum target to interference function level and contrast difference at the target cluster center,

means for developing said target cluster size and maximum target to interference function from said target to interference related function, and

means for creating said contrast difference at target cluster center from said target to interference related function and said average contrast difference.

22. A target detection method as defined in claim 21, wherein said means for determining the local grey level variation comprises

averaging the vertical picture element values,

averaging the horizontal picture element values, and

determining a grey level variation from said vertical picture element value and said horizontal picture element value.

23. A target detection method as defined in claim 15, wherein said MLC step comprises

developing an adaptive threshold level for target windows, and

producing an edge direction map for each picture element in a target sized window.

24. A target detection method as defined in claim 23, wherein said step for developing an adaptive threshold level-for target windows comprises

determining the intensity gradient for non-overlapping neighborhoods having less than target size areas,

determining the average intensity gradient for the entire image area,

a look-up table step for developing an average global gradient from said average entire image area gradient,

averaging the intensity gradient of a group of said non-overlapping neighborhoods and,

estimating the standard deviation of said group of said non-overlapping neighborhoods,

summing a function of said standard deviation of said group of said non-overlapping neighborhoods with said average intensity gradient as a group of said non-overlapping neighborhoods,

a look-up table step for developing a function of said standard deviation of said group of said non-overlapping neighborhoods, and

producting said average global gradient, said average intensity gradient of said group of said non-overlapping neighborhoods, and said look-up table developed function of said standard deviation of said group of said non-overlapping neighborhoods.

25. A target detection method as defined in claim 23 wherein said step for producing an edge direction map for each picture element in a target sized window comprises:

a convolutional step for generating a joint edge map from said edge directional map data,

computing histograms for target sized window increments of data for those target sized windows containing data above said adaptive threshold level from said joint edge map data and said gradient intensity data,

said histogram step including extracting lower, middle, and upper classes of feature vectors from data representing each selected target window, said each class of feature vectors bounded by upper and lower values and having overlapping ranges,

a lower, middle, and upper look-up table step for mapping said feature vectors as probability of target functions, and

producing a target affirmative indication when the probability is 0.5 or greater and a non-target indication when the probability is less than 0.5.

26. A target detection method as defined in claim 15, wherein said VSC step comprises

developing a grey level histogram of a target sized window,

calculating upper and lower threshold levels for said histogram as a function of histogram characteristics,

eliminating picture elements falling below the low level value,

calculating the centroid of a cluster of pixels exceeding said threshold value,

creating a global histogram from the entire video image,

calculating a first optimal global histogram from said global histogram and said upper grey level threshold,

calculating a second global histogram function from said lower grey level threshold,

calculating a target sized histogram from said target sized histogram data and said low level threshold value, and

a look-up table step responsive to said upper threshold level global histogram said lower threshold level global histogram said target sized window histogram and said centroid for producing a probability of target function and a confidence function.

27. A target detection method as defined in claim 15, wherein said HPL step comprises

determining if an input data cluster contains a probability of target assessment,

determining the types of probability assessments,

forwarding MLC probability of target data to a one dimensional decision Parzen table,

forwarding VSC data to a one dimensional decision Parzen table,

merging MLC probability target and VSC probability of target and data,

a two dimensional Parzen table step to said merged MLC and VSC data,

merging MLC probability of target and GTIR probability of target and data,

providing a two feature Parzen table responsive to said merged MLC and GTIR data,

providing a single feature Parzen table responsive to GTIR probability of target data,

merging GTIR probability of target and VSC probability of target and data,

providing a two feature Parzen table responsive to said merged GTIR and VSC probability of target data,

merging MLC, GTIR, and VSC probability of target, and

providing a three feature Parzen table responsive to said merged MLC, GTIR, and VSC probability of target data.

28. The target detection method as in claim 15, in which said combining step is carried out in a joint decision space.

29. The target detection method as in claim 15, further including

running a training set of known images through each of the above four steps to train the system.

30. Target detection apparatus comprising

first means for determining first feature data and for mapping the same into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from a training set of known examples of known imagery to give a first list of targets,

second means for determining second ratio matched feature data in such a manner so to be independent or anticorrelated to said first means, and for mapping the same into a probability of target with confidence bounds based a maximum likelihood statistic experimentally derived from said training set to give a second list of targets,

said first and second feature data being statistically independent or anticorrelated,

means for combining or mapping the target lists from said first and second means into a composite probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from said training set, and

target prioritization means for ranking the target output of said means to yield a target ranking list.

31. In a target detection method for detecting targets in a sensed image to determine the correlation probability of objects in the image to stored feature recognition data, comprising

generating a feature pattern corresponding to a first algorithmic point of view of the scene,

matching the feature pattern with a stored set of feature recognition data of the same algorithmic point of view to determine the correlation probability therewith,

generating one or more other feature patterns from one or more other points of view of the scene which are independent of the first point of view in all combinations, both singly and jointly,

matching the other feature patterns with other stored sets of feature recognition data of the other points of view,

said points of view being represented by algorithms different from each other in such ways as to generate correlated true detection but independent of anticorrelated false detections in all combinations, both singly and jointly.

32. Apparatus for detecting targets in a sensed image to determine the correlation probability of objects in the image to stored feature recognition data, comprising

means for generating a feature pattern corresponding to a first algorithmic point of view of the scene,

means for matching the feature pattern with a stored set of feature recognition data of the same algorithmic point of view to determine the correlation probability therewith,

means for generating one or more other feature patterns from one or more other points of view of the scene which are independent of the first point of view in all combinations, both singly and jointly,

means for matching the other feature patterns with other stored sets of feature recognition data of the other points of view,

said points of view being represented by algorithms different from each other in such ways as to generate correlated true detection but independent of anticorrelated false detections in all combinations, both singly and jointly.

33. A target detection method comprising

determining joint edge events and for mapping the same into a measure of probability of target based on a maximum likelihood junction derived from a training set,

determining target-to-interference ratio data and for mapping the same into a measure of probability of target based on a matched filter function,

determining spatial clustering features and for mapping the same into a measure of probability of target based on a geometric function,

said MLC, GTIR, VSC, and HPL steps being capable of being trained by a set of known targets in sets of exemplar visual fields to develop a knowledge base,

said MLC, GTIR, and VSC steps being substantially singly and jointly independent of or anticorrelating of each other in all combinations,

HPL combining the target detection outputs from said MLC, GTIR, and VSC means into a mapping function based upon the training data,

so that said system is ready to process and identify unknown targets in new visual fields,

running a training set including known targets each of the above process steps to develop a knowledge base of extrapolated known situations subsequently processing new visual fields with unknown targets using said knowledge base through each of said MLC, GTIR, VSC, and HPL steps.

34. A target detection system comprising

MLC means for determining joint edge events and for mapping the same into a measure of probability of target based on a maximum likelihood function derived from a training set to develop a knowledge base,

GTIR means for determining target-to-interference ratio data and for mapping the same into a measure of probability of target based on a matched filter function derived from a training set to develop a knowledge base,

VSC means for determining spatial clustering features and for mapping the same into a measure of probability of target based on a geometric function derived from a training set to develop a knowledge base,

said MLC, GTIR, and VSC means being substantially singly and jointly independent of or anticorrelating of each other in all combinations,

HPL means for combining the target detection outputs from said MLC, GTIR, and VSC means into a mapping function based upon the training data, said MLC, GTIR, VSC, and HPL means being capable of being trained by a set of known targets in sets of visual fields to develop a knowledge base,

so that said system is ready to process and identify unknown targets in new visual fields.

35. Target detection apparatus comprising

MLC means for determining joint edge feature data and for mapping the same into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from a training set of known examples of known imagery to select a target,

GTIR means for determining target-to-interference ratio matched filter feature data in such a manner so to be independent or anticorrelated to said MLC means, and for mapping the same into a probability of target with confidence bounds based on a maximum likelihood statistic experimentally derived from said training set to select a target, and

means responsive to said MLC means for aiming a missile at said target.

36. A target detection apparatus comprising

GTIR means for determining target-to-interference ratio matched filter feature data in such a manner so to be independent or anticorrelated to other target detection means, and for mapping the same into a probability of target with confidence bounds based on a maximum likelihood statistic experimentally derived from said training set to select a target,

said GTIR means including

means for determining the average local contrast difference between target related picture elements and background related picture elements,

means for determining the local grey level variation, and

means for generating a target to interference related function from said local contrast difference and said local grey level variations.

37. A target detection apparatus as defined in claim 36, comprising

table look-up means for generating a probability of target function and a confidence function from said target to interference function and said local contrast difference

said table look-up means comprises a three feature Parzen table array,

means for generating a feature vector input to said three feature Parzen table from the cluster size, maximum target to interference function level and contrast difference at the target cluster center,

means for developing said target cluster size and maximum target to interference function from said target to interference related function, and

means for creating said contrast difference at target cluster center from said target to interference related function and said average contrast difference.

38. Target detection apparatus comprising

MLC means for determining joint edge feature data and for mapping the same into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from a training set of known examples of known imagery to select a target, and

means responsive to said MLC means for aiming a missile at said target.

39. A target detection apparatus as defined in claim 38, wherein said MLC means comprises

means for developing an adaptive threshold level for target windows, and

means for producing an edge direction map for each picture element in a target sized window.

40. A target detection apparatus as defined in claim 39, wherein said means for developing an adaptive threshold level for target windows comprises

means for determining the intensity gradient for non-overlapping neighborhoods having less than target size areas,

means for determining the average intensity gradient for the entire image area,

look-up table means for developing an average global gradient from said average entire image area gradient,

means for averaging the intensity gradient of a group of said non-overlapping neighborhoods and,

means for estimating the standard deviation of said group of said non-overlapping neighborhoods,

means for summing a function of said standard deviation of said group of said non-overlapping neighborhoods with said average intensity gradient as a group of said non-overlapping neighborhoods,

look-up table means for developing a function of said standard deviation of said group of said non-overlapping neighborhoods, and

means for producting said average global gradient, said average intensity gradient of said group of said non-overlapping neighborhoods, and said look-up table developed function of said standard deviation of said group of said non-overlapping neighborhoods.

41. A target detection apparatus as defined in claim 39 wherein said means for producing an edge direction map for each picture element in a target sized window comprises:

convolutional means for generating a joint edge map from said edge directional map data,

means for computing histograms for target sized window increments of data for those target sized windows containing data above said adaptive threshold level from said joint edge map data and said gradient intensity data,

said histogram including means for extracting lower, middle, and upper classes of feature vectors from data representing each selected target window, said each class of feature vectors bounded by upper and lower values and having overlapping ranges,

lower, middle, and upper look-up table means for mapping said feature vectors as probability of target functions, and

means for producing a target affirmative indication when the probability is 0.5 or greater and a non-target indication when the probability is less than 0.5.

42. In target detection apparatus including a plurality of means for target selection

HPL means for combining said target selections from said means into a composite probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from a training set, and

target prioritization means for ranking the target output of said HPL means to yield a target ranking list.

43. A target detection apparatus comprising

MLC means for determining joint feature data and for mapping the same into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from a training, of known examples of known imagery to select a target,

GTIR means for determining target-to-interference ratio matched filter feature data in such a manner so to be independent or anticorrelated to other target detection means, and for mapping the same into a probability of target with confidence bounds based on a maximum likelihood statistic experimentally derived from said training set to select a target,

VSC means for determining grey level feature data in such a manner as to be independent or anticorrelated singly and jointly to said MLC and GTIR means, and for mapping said grey level feature data into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from said training set to select a target,

HPL means for combining (or mapping) said target selections from said means into a composite probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from a training set,

means for determining if an input data cluster contains a probability of target assessment,

means for determining the types of probability assessments,

means for forwarding MLC probability of target data to a one dimensional decision Parzen table,

means for forwarding VSC data to a one dimensional decision Parzen table,

means for merging MLC probability target and VSC probability of target and data,

a two dimensional Parzen table responsive to said merged MLC and VSC data,

means for merging MLC probability of target and GTIR probability of target and data,

a two feature Parzen table responsive to said merged MLC and GTIR data,

a single feature Parzen table responsive to GTIR probability of target data,

means to merge GTIR probability of target and VSC probability of target and data,

a two feature Parzen table responsive to said merged GTIR and VSC probability of target data,

means to merge MLC, GTIR, and VSC probability of target, and

a three feature Parzen table responsive to said merged MLC, GTIR, and VSC probability of target data, and

target prioritization means for ranking the target output of said HPL means to yield a target ranking list.

44. Apparatus as defined in claim 43, comprising

final target selection means for ranking the outputs of said decision Parzen tables representing the entire field of view,

a dual mode tracker responsive to the outputs of said final target selection means for selecting a desired target,

said dual mode tracker including a minimum absolute difference correlation tracker 27 and a centroid tracker,

sensor formatter for digitizing the field of view containing the selected target and its immediate surroundings as a target template,

means for generating a minimum absolute difference function for the field of view represented by said template,

means for compensating for tracking vehicle motion,

means for calculating the correlation between a sensed image and a reference provided by said template, and

means for providing feedback signals to the vehicle direction control system.

45. Target detection apparatus comprising

MLC means for determining joint edge feature data and for mapping the same into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from a training set of known examples of known imagery, and

VSC means for determining grey level feature data in such a manner as to be independent or anticorrelated singly and jointly to said MLC means, and for mapping said grey level feature data into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from said training set,

means for combining or mapping the target detection outputs from MLC and VSC into a ranked composite probability to select a most probable target.

46. Target detection method comprising

MLC means for determining joint edge feature data and for mapping the same into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from a training set of known examples of known imagery to select a target,

a GTIR step for determining target-to-interference ratio matched filter feature data in such a manner so to be independent or anticorrelated to said MLC means, and for mapping the same into a probability of target with confidence bounds based on a maximum likelihood statistic experimentally derived from said training set to select a target, and

a step responsive to the output of one of said MLC or GTIR step for aiming a missile at said target.

47. A target detection method comprising

a GTIR step for determining target-to-interference ratio matched filter feature data in such a manner so to be independent or anticorrelated to said MLC means, and for mapping the same into a probability of target with confidence bounds based on a maximum likelihood statistic experimentally derived from said training set to select a target,

determining the average local contrast difference between target related picture elements and background related picture elements,

determining the local grey level variation, and

generating a target to interference related function from said local contrast difference and said local grey level variations.

48. A target detection system as defined in claim 47 comprising

generating a probability of target function and a confidence function from said target to interference function and said local contrast difference and storing the same in a look-up table comprising a three feature Parzen table array,

generating a feature vector input to said three feature Parzen table array from the cluster size maximum target to interference function level and contrast difference at the target cluster center,

developing said target cluster size and maximum target to interference function from said target to interference related function, and

creating said contrast difference at target cluster center from said target to interference related function and said average contrast difference.

49. A target detection method as defined in claim 48, wherein said means for determining the local grey level variation comprises

averaging the vertical picture element values,

averaging the horizontal picture element values, and

determining a grey level variation from said vertical picture element value and said horizontal picture element value.

50. Target detection apparatus comprising

an MLC step for determining joint edge feature data and for mapping the same into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from a training set of known examples of known imagery to select a target, and

a step responsive to said MLC step for aiming a missile at said target.

51. A target detection method as defined in claim 50, wherein said MLC step comprises

developing an adaptive threshold level for target windows, and

producing an edge direction map for each picture element in a target sized window.

52. A target detection method as defined in claim 51 wherein said step for developing an adaptive threshold level for target windows comprises

determining the intensity gradient for non-overlapping neighborhoods having less than target size areas,

determining the average intensity gradient for the entire image area,

a look-up table step for developing an average global gradient from said average entire image area gradient,

averaging the intensity gradient of a group of said non-overlapping neighborhoods,

estimating the standard deviation of said group of said non-overlapping neighborhoods,

summing a function of said standard deviation of said group of said non-overlapping neighborhoods with said average intensity gradient as a group of said non-overlapping neighborhoods,

a look-up table step for developing a function of said standard deviation of said group of said non-overlapping neighborhoods, and

producting said average global gradient, said average intensity gradient of said group of said non-overlapping neighborhoods, and said look-up table developed function of said standard deviation of said group of said non-overlapping neighborhoods.

53. A target detection method as defined in claim 51 wherein said step for producing an edge direction map for each picture element in a target sized window comprises

a convolutional step for generating a joint edge map from said edge directional map data,

computing histograms for target sized window increments of data for those target sized windows containing data above said adaptive threshold level from said joint edge map data and said gradient intensity data,

said histogram step including extracting lower, middle, and upper classes of feature vectors from data representing each selected target window, said each class of feature vectors bounded by upper and lower values and having overlapping ranges,

a lower, middle, and upper look-up table step for mapping said feature vectors as probability of target functions, and

producing a target affirmative indication when the probability is 0.5 or greater and a non-target indication when the probability is less than 0.5.

54. Target detection method comprising

a MLC step for determining joint edge feature data and for mapping the same into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from a training set of known examples of known imagery to select a target,

a VSC step for determining grey level feature data in such a manner as to be independent or anticorrelated singly and jointly to said MLC step, and for mapping said grey level feature data into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from said training set to select a target, and

a step responsive to one of said MLC step or VSC for aiming a missile at said target.

55. A target detection method comprising

a VSC step for determining grey level feature data, and for mapping said grey level feature data into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from said training set to select a target, including the steps of

developing a grey level histogram of a target sized window,

calculating upper and lower threshold levels for said histogram as a function of histogram characteristics,

eliminating picture elements falling below the low level value,

calculating the centroid of a cluster of pixels exceeding said threshold value,

creating a global histogram from the entire video image,

calculating a first optimal global histogram from said global histogram and said upper grey level threshold,

calculating a second global histogram function from said lower grey level threshold,

calculating a target sized histogram from said target sized histogram data and said low level threshold value, and

a look-up table step responsive to said upper threshold level global histogram said lower threshold level global histogram said target sized window histogram and said centroid for producing a probability of target function and a confidence function.

56. Method for detecting targets in a sensed image to determine the correlation probability of objects in the image to stored feature recognition data, comprising

generating a feature pattern corresponding to a first algorithmic point of view of the scene,

matching the feature pattern with a stored set of feature recognition data of the same algorithmic point of view to determine the correlation probability therewith,

generating one or more other feature patterns from one or more other points of view of the scene which are independent of the first point of view in all combinations, both singly and jointly,

matching the other feature patterns with other stored sets of feature recognition data of the other points of view, said points of view being represented by algorithms different from each other in such ways as to generate correlated true detection but independent of anticorrelated false detections in all combinations, both singly and jointly.

57. Target detection method comprising the steps of:

determining a first feature data and mapping the same into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from a training set of known examples of known imagery,

determining a second feature data in such a manner so to be independent or anticorrelated to said first data and mapping the same into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from said training set, and

combining or mapping the target detection outputs from said first and second feature data determining steps into a composite probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from said training set.

58. Target detection apparatus comprising

MLC means for determining joint edge feature data and for mapping the same into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from a training set of known examples of known imagery to select a target,

VSC means for determining grey level feature data in such a manner as to be independent or anticorrelated singly and jointly to said MLC and GTIR means, and for mapping said, grey level feature data into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from said training set to select a target, and

means responsive to one of said MLC or VSC means for aiming a missile at said target.

59. A target detection apparatus comprising

VSC means for determining grey level feature data and for mapping grey level feature data into a probability of target with confidence bounds based upon a maximum likelihood statistic experimentally derived from said training set to select a target,

said VSC means including

means for developing a grey level histogram of a target sized window,

means for calculating upper and lower threshold levels for said histogram as a function of histogram characteristics,

means for eliminating picture elements falling below the low level value,

means for calculating the centroid of a cluster of pixels exceeding said threshold value,

means for creating a global histogram from the entire video image,

means for calculating a first optimal global histogram from said global histogram and said upper grey level threshold,

means for calculating a second global histogram function from said lower grey level threshold,

means for calculating a target sized histogram from said target sized histogram data and said low level threshold value, and

look-up table means responsive to said upper threshold level global histogram said lower threshold level global histogram said target sized window histogram and said centroid for producing a probability of target function and a confidence function.