Moving target detection method using two-frame subtraction and a two quadrant multiplier
First Claim
1. An apparatus for detecting a moving object in a cluttered scene comprising:
- first means for storing data of a first representation of the cluttered scene;
second means for storing data of a second representation of the cluttered scene;
means for subtracting the first representation from the second representation to form a subtracted representation of the scene;
a first spatial filter for filtering the subtracted representation;
a second spatial filter for filtering the data in the first means for storing; and
a two quadrant multiplier, said two quadrant multiplier receiving an output from said first spatial filter and an output from said second spatial filter to form an output substantially eliminating the background clutter and an ambiguous image of the object stored in the second means for storing,
3 Assignments
0 Petitions
Accused Products
Abstract
A method and apparatus for detecting an object of interest against a cluttered background scene. In a first preferred embodiment the sensor tracking the scene is movable on a platform such that each frame of the video representation of the scene is aligned, i.e., appears at the same place in sensor coordinates. A current video frame of the scene is stored in a first frame storage device (14) and a previous video frame of the scene is stored in a second frame storage device (20). The frames are then subtracted by means of an invertor (24)and a frame adder (28) to remove most of the background clutter. The subtracted image is put through a first leakage reducing filter, preferably a minimum difference processor filter (32). The current video frame in the first frame storage device (14) is put through a second leakage-reducing filter, preferably minimum difference processor filter (36). The outputs of the two processors are applied to a two quadrant multiplier (42) to minimize the remaining background clutter leakage and to isolate the moving object of interest. In a second preferred embodiment the sensor does not track the scene and therefore the frames of video from the two storage devices (140, 200) must be aligned before the subtraction process can occur. The alignment function is performed by a background registration correlator (150) and a frame shifter (130). Once the two frames of video data are aligned they are then subtracted and preferably applied to the minimum difference processor filters (320, 360) as with the first embodiment. The outputs of the two minimum difference processors (320, 360) are again applied to a two quadrant multiplier (420) to eliminate the remaining background clutter.
-
Citations
24 Claims
-
1. An apparatus for detecting a moving object in a cluttered scene comprising:
-
first means for storing data of a first representation of the cluttered scene; second means for storing data of a second representation of the cluttered scene; means for subtracting the first representation from the second representation to form a subtracted representation of the scene; a first spatial filter for filtering the subtracted representation; a second spatial filter for filtering the data in the first means for storing; and a two quadrant multiplier, said two quadrant multiplier receiving an output from said first spatial filter and an output from said second spatial filter to form an output substantially eliminating the background clutter and an ambiguous image of the object stored in the second means for storing, - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
-
-
10. A method for detecting a moving object in a cluttered scene comprising the steps of:
-
storing a first representation of the cluttered scene; storing a second representation of the cluttered scene; subtracting the second representation from the first representation to form a subtracted representation; filtering the subtracted representation through a first spatial filter; filtering the first representation in a second nonlinear spatial filter; and multiplying outputs from the first and second spatial filters in a two quadrant multiplier to detect the current image of the object and eliminate an ambiguous image of the object. - View Dependent Claims (11, 12, 13, 14, 15, 16, 17)
-
-
18. A method for detecting a moving object in a cluttered scene comprising the steps of:
-
storing a first frame of video data of the cluttered scene at a first time in a first frame storage device; transferring the first frame of video data from the first storage device to a second frame storage device at a second time; storing a second frame of video data of the cluttered scene in the first frame storage device substantially at the second time; subtracting the second frame of video data from the first frame of video data to form subtracted video data; filtering the subtracted video data in a first spatial filter; filtering the first frame of video data through a second spatial filter; and multiplying outputs from the first and second spatial filters in a two quadrant multiplier to detect the object and eliminate an ambiguous image of the object. - View Dependent Claims (19, 20, 21, 22, 23, 24)
-
Specification