초록 ▼

A land-based smart sensor system and several system architectures for detection, tracking, and classification of people and vehicles automatically and in real time for border, property, and facility security surveillance is described. The preferred embodiment of the proposed smart sensor system is c

A land-based smart sensor system and several system architectures for detection, tracking, and classification of people and vehicles automatically and in real time for border, property, and facility security surveillance is described. The preferred embodiment of the proposed smart sensor system is comprised of (1) a low-cost, non-coherent radar, whose function is to detect and track people, singly or in groups, and various means of transportation, which may include vehicles, animals, or aircraft, singly or in groups, and cue (2) an optical sensor such as a long-wave infrared (LWIR) sensor, whose function is to classify the identified targets and produce movie clips for operator validation and use, and (3) a supercomputer to process the collected data in real-time. The smart sensor system can be implemented in a tower-based or a mobile-based, or combination system architecture. The radar can also be operated as a stand-alone system.

대표청구항 ▼

1. A method for detection of the presence of at least one target on land using a non-coherent radar, comprising: (a) transmitting a plurality of radar pulses by said non-coherent radar toward an area of interest;(b) receiving a radar return scattered from said area of interest, said radar return inc

1. A method for detection of the presence of at least one target on land using a non-coherent radar, comprising: (a) transmitting a plurality of radar pulses by said non-coherent radar toward an area of interest;(b) receiving a radar return scattered from said area of interest, said radar return including a return from said target and interference from system noise and background noise and clutter; and(c) processing said radar return for at least one range-azimuth resolution cell in said area of interest to differentiate said return from said target from said interference from said system noise and background noise and clutter, said processing including: applying at least one threshold to differentiate said target from said noise and system noise and said background clutter;comparing the radar return to a first threshold of the at least one threshold, the first threshold set at a level to identify multiple radar contacts including the target and false contacts from said interference from system noise and background noise and clutter, wherein a radar contact is declared when said radar return exceeds said first threshold; andtracking the multiple radar contacts over time to distinguish the target from said false contacts. 2. The method of claim 1, wherein said target moves through said at least one resolution cell. 3. The method of claim 1, wherein said processing includes processing of said radar return for a plurality of resolution cells in which said return is associated with said target. 4. The method of claim 3, wherein said processing step includes processing of said radar return for a plurality of neighboring resolution cells. 5. The method of claim 4, wherein said processing step includes processing of said radar return for a plurality of proximate resolution cells. 6. The method of claim 1, wherein the radar cross-section of said target is less than or equal to one square meter. 7. The method of claim 1, wherein said radar transmits pulses throughout an angle of 360 degrees. 8. The method of claim 7, wherein said radar pulses are transmitted in a plurality of separate, adjacent and independent radar sectors. 9. The method of claim 8, wherein said sectors overlap. 10. The method of claim 1, wherein said target of interest is not moving. 11. The method of claim 10, wherein said processing includes comparing a first radar return obtained with said target present in one of said at least one resolution cell to a second radar return obtained in said one of said at least one resolution cell without said target present. 12. The method of claim 1, wherein a statistical distribution derived from said radar return is used to set one of said at least one threshold with a user-designated probability of detection and a user-designated probability of false alarm. 13. The method of claim 12, wherein said processing said return includes determining the presence of said target from a decision criterion comprised of an accumulation of one or more threshold exceedances in one or more resolution cells in close proximity to each other that is consistent with (1) the expected intensity of said radar return from an actual target and (2) the expected movement of an actual target. 14. The method of claim 1, wherein a contact is declared when said return exceeds said at least one threshold and wherein said processing said return includes an empirically-based method to distinguish false-contacts due to said noise and clutter from actual contacts due to said target by using noise and clutter obtained at a different time when a target is not present. 15. The method of claim 1, wherein said target is moving and wherein said processing said return includes a track-before-detect method. 16. The method of claim 15, wherein a contact is declared when said return exceeds said at least one threshold and wherein said track-before-detect method comprises: (a) accumulating contacts in each resolution cell in said area of interest by setting a threshold for each radar measurement across said area of interest that includes both actual contacts due to said targets and false contacts due to said noise and clutter; and(b) determining the presence of said targets from a comparison of tracks of said contacts with expected tracks of said actual contacts and said false contacts in one or more resolution cells in close proximity to each other. 17. The method of claim 1, wherein said at least one target is selected from the group consisting of people, vehicles, and animals. 18. The method of claim 1, wherein said non-coherent radar is an X-Band radar. 19. The method of claim 1, wherein said non-coherent radar is a Ku-Band radar. 20. The method of claim 1, wherein said non-coherent radar is a C-Band radar. 21. The method of claim 1, wherein said detection is performed using at least two radars located at different places. 22. The method of claim 1, wherein said target is tracked over a period of time. 23. The method of claim 22, wherein said target is moving in an environment in which said target is not detectable by said radar at all target locations. 24. The method of claim 23, wherein said target is not within a line-of-sight of said radar at all times during said period of time. 25. The method of claim 1, wherein a detection of a moving target is declared based on the development of a track. 26. The method of claim 25, wherein said track is determined from a probability statistic. 27. The method of claim 25, wherein the speed of said target along said track is consistent with that of an actual target. 28. The method of claim 25, wherein the trajectory of said track is consistent with that of an actual target. 29. The method of claim 25, wherein a contact is declared when said return exceeds said at least one threshold and wherein said track is consistent with an expected track of said target. 30. The method of claim 1, wherein said non-coherent radar is also used for tracking or classification of said target. 31. The method of claim 30, wherein classification is based on target characteristics including characteristic speed, movement, trajectory, and intensity of said radar return. 32. The method of claim 31, wherein classification is further based on a statistical distribution of said target characteristics. 33. The method of claim 1, wherein said processing said return includes applying an algorithm that determines whether a track derived from said radar return for at least two time intervals is consistent with an expected track for an actual target. 34. The method of claim 33, wherein said track is determined by the level of confidence of a probability statistic applied to said radar return. 35. The method of claim 34, wherein said probability statistic is defined by a log-likelihood ratio with the assumption that said system noise and clutter within each one of said at least one resolution cell is substantially constant over a period of time, and wherein target-absent statistics over multiple radar scans is determined and used in the determination of the log-likelihood ratio. 36. The method of claim 34, wherein said probability statistic is based on Bayesian statistical methods. 37. The method of claim 1, wherein said receiving includes collecting and processing at least two samples of said radar returns from said at least one range-azimuth resolution cell. 38. The method of claim 1, wherein said processing said radar return includes pre-processing data from said radar return to generate a detection field before said return is differentiated. 39. The method of claim 38, wherein said detection field is generated after conditioning operations are performed on said return, wherein said conditioning operations include an operation selected from the group consisting of calibration, spatial registration, pre-filtering, constant background removal, and noise background estimation. 40. The method of claim 39, wherein said spatial registration includes aligning resolution cells for direct comparison from one scan to the next scan. 41. The method of claim 38, wherein said detection field is generated by segmentation of said data. 42. The method of claim 38, wherein said pre-processing includes separating resolution cells into different categories for processing based on the type of noise and clutter present in said resolution cells. 43. The method of claim 42, wherein said pre-processing includes separating resolution cells dominated by system/receiver noise. 44. The method of claim 1, wherein a second threshold is used to differentiate false contacts due to said noise and clutter from actual targets, said second threshold being based on a track-before-detect method. 45. The method of claim 1, wherein a contact is declared when said return exceeds said at least one threshold, wherein said receiving includes receiving the radar return from multiple scans of said radar over said area of interest collected over a period of time, and wherein said processing includes applying two thresholds: a first threshold set to detect a first set of contacts, said first set of contacts including actual contacts as well as false contacts, and a second threshold to differentiate said actual contacts from said false contacts, and wherein said processing includes applying an odds ratio to said first set of contacts, said odds ratio being configured to increase when an actual contact is present until it exceeds said second threshold. 46. The method of claim 45, wherein said odds ratio is computed based on Bayes Rule, said odds ratio being multiplied by a likelihood ratio at each observation to update said odds ratio. 47. The method of claim 1, further comprising: (a) determining background statistics of said radar return for each of said at least one resolution cell when a target is not present;(b) using said background statistics to remove said noise and clutter from said radar return from said area of interest;(c) detecting a target when said radar return with noise and clutter removed, obtained in (b), exceeds a threshold value. 48. The method of claim 1, wherein said non-coherent radar cues an optical sensor system to image said target. 49. The method of claim 48, wherein said optical sensor system is configured to detect, track, or classify said target. 50. The method of claim 49, wherein classification is based on at least one target characteristic selected from the group consisting of speed, movement, trajectory, and image intensity. 51. The method of claim 48, wherein said non-coherent radar is configured to track or classify said target. 52. The method of claim 48, further comprising a plurality of sensor systems, including non-coherent radar and optical sensors, located at different positions, said plurality of sensor systems being configured in conjunction with said non-coherent radar and said optical sensor system for detection, tracking or classification of said target. 53. The method of claim 48, wherein an image of said target is sent to a person for validation of the existence and type of target present. 54. The method of claim 48 further comprising processing pixels in the image of said target to differentiate said target from system and background noise, said processing including applying at least one optical threshold to differentiate said target from said noise. 55. The method of claim 54, wherein a statistical distribution derived from said pixels in the image is used to set a threshold with a user-designated probability of detection and a user-designated probability of false alarm. 56. The method of claim 55, wherein said processing said pixels includes determining the presence of said target from a decision criterion comprised of an accumulation of one or more threshold exceedances in one or more resolution cells in close proximity to each other that is consistent with (1) the expected intensity of the radar return from an actual target and (2) the expected movement of an actual target. 57. The method of claim 54, wherein a contact is declared when the processed pixels in the image exceed said at least one threshold and wherein said processing pixels includes an empirically-based method to distinguish false contacts due to said noise from actual contacts due to said target by using noise obtained at a different time when a target is not present. 58. The method of claim 54, wherein said target is moving and wherein said processing pixels includes a track-before-detect method. 59. The method of claim 58, wherein a contact is declared when the processed pixels in said image exceed said at least one threshold and wherein said track-before-detect method comprises: (a) accumulating contacts in each pixel in said area of interest by setting a threshold for each pixel of said area of interest that includes both actual contacts due to said targets and false contacts due to said noise; and(b) determining the presence of said targets from a comparison of tracks of said contacts with expected tracks of said actual contacts and said false contacts in one or more pixels in close proximity to each other. 60. The method of claim 48, further comprising: (a) determining background statistics of each pixel of the optical sensor image obtained when a target is not present;(b) using said background statistics to remove background and environmental noise in said optical sensor image; and(c) detecting a target when said optical sensor image with background and environmental noise removed, obtained in (b), exceeds a pre-defined threshold level. 61. The method of claim 1, wherein said non-coherent radar cues a coherent radar sensor system to detect said target. 62. The method of claim 61, wherein said coherent radar sensor system is configured to detect, track, or classify said target. 63. The method of claim 61, further including an optical sensor system, wherein said non-coherent radar cues said optical sensor system to image said target. 64. The method of claim 1, wherein the radar cross-section of said target is less than or equal to the radar cross-section of a human. 65. The method of claim 1, wherein the radar cross-section of said target is less than or equal to the radar cross-section of a group of two humans.