Vehicular target acquisition and tracking using a generalized hough transform for missile guidance
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G06K-009/00
F41G-007/00
출원번호
US-0858741
(2004-06-02)
등록번호
US-7444002
(2008-10-28)
발명자
/ 주소
Teku,Hagos
Lo,Thomas K.
Lawrence,Nikki J.
출원인 / 주소
Raytheon Company
대리인 / 주소
Noblitt & Gilmore, LLC
인용정보
피인용 횟수 :
4인용 특허 :
5
초록▼
A generalized Hough transform is used to acquire and track vehicular targets for missile guidance. This is accomplished by recognizing that most vehicles have silhouettes that may be described as a "rounded rectangle". The position and shape of such rounded rectangles is described in terms of 5 para
A generalized Hough transform is used to acquire and track vehicular targets for missile guidance. This is accomplished by recognizing that most vehicles have silhouettes that may be described as a "rounded rectangle". The position and shape of such rounded rectangles is described in terms of 5 parameters (xc, yc, a, b, θ) where xc, yc are the center coordinates, a,b are the major and minor axis and Θ is the orientation. The computation of a five dimensional Hough transform on an image including such a target will produce the five parameters that provide the "best fit" rounded rectangle to the target. These parameters are then passed to a missile tracker. This capability can be used to improve track gate handoff from the automatic target recognizer (ATR) to the missile tracker, changing aspect ratios of maneuvering targets, limited lock-on after launch (LOAL), aimpoint designation and fire control system to missile seeker handover.
대표청구항▼
We claim: 1. A method, comprising: acquiring an image of a scene containing a vehicle; performing edge detection on the image to produce a silhouette image; performing a Hough transform with parameters including center coordinates xc, yc, major and minor axes a,b, and orientation Θ on the silh
We claim: 1. A method, comprising: acquiring an image of a scene containing a vehicle; performing edge detection on the image to produce a silhouette image; performing a Hough transform with parameters including center coordinates xc, yc, major and minor axes a,b, and orientation Θ on the silhouette image to detect a best fit of a parameterized object to the vehicle; and passing position, size and orientation parameters of the object to a missile tracker. 2. The method of claim 1, wherein the parameterized object is a rounded rectangle. 3. The method of claim 2, wherein the rounded rectangle is given by: wherein the vehicle lies along the x and y axes with parameters including center coordinates xc, yc, major and minor axes a,b, and orientation Θ. 4. The method of claim 3, wherein exponent n assumes a range of values so that the rounded rectangle ranges from an ellipse to a rectangle. 5. The method of claim 3, wherein exponent n lies between 2 and 3 inclusive. 6. The method of claim 3, wherein exponent n lies between 2.1 and 2.6 inclusive. 7. The method of claim 3, wherein exponent n is selected by offline analysis of a collection of representative sample vehicles with varying aspects and parametrically varying the value n to arrive at a value which best fits the samples. 8. The method of claim 1, wherein the Hough transform is computed as a sequence of 2D Hough transforms by fixing three of the parameters (xc, yc, a, b, θ), computing a 2D Hough transform for the other two parameters, and iterating on the three fixed parameters. 9. The method of claim 1, wherein the Hough transform is normalized by the perimeter of the parameterized object to determine the best fit. 10. The method of claim 1, further comprising: placing an initial track suite that does not conform to the vehicle over the vehicle, said edge detection being performed on the portion of the image within the track gate to produce the silhouette image; adjusting the position and size of the track gate based on said position, size and orientation parameters so that said track gate is approximately conformal with the vehicle; and passing the conformal track gate to the missile tracker. 11. The method of claim 10, wherein the initial track gate is acquired from an automatic target recognizer (ATR) on the missile and the missile tracker's target acquisition module performs the edge detection and Hough transform to adjust the track gate. 12. The method of claim 11, wherein the missile tracker's target acquisition module periodically updates the parameters to adjust the track gate to account for aimpoint drift during tracking or vehicle maneuvering. 13. The method of claim 11, wherein the image is acquired with a high resolution off-missile imager and the parameters are computed with an off-missile target acquisition module and passed to the missile having a low resolution imager with a different line-of-sight than the high resolution imager, said missile tracker's target acquisition module recomputing the Hough transform in a constrained parameter space based on the parameters to estimate the line-of-sight difference between the missile and the off-missile imager. 14. The method of claim 10, wherein the initial track gate is positioned on or near the vehicle by a laser designator. 15. The method of claim 1, further comprising selecting an aimpoint on the vehicle based on the position, size and orientation parameters. 16. A method, comprising: acquiring an image of a scene containing a vehicle with a high resolution off-missile imager; performing edge detection on the image to produce a silhouette image; performing a Hough transform on the silhouette image to detect a best fit of a parameterized object to the vehicle; and passing position, size and orientation parameters of the object from the high resolution imager to a missile seeker on a missile over a low bandwidth channel, said missile seeker including a low resolution imager that acquires low resolution images of the scene with a different line-of-sight than the high resolution imager and a missile tracker that recomputes the Hough transform on the low resolution images in a constrained spaced based on the parameters to compensate for the different line-of-sight. 17. A method, comprising: acquiring an image of a scene containing a vehicle; performing edge detection on the image to produce a silhouette image; performing a Hough transform on the silhouette image to detect a best fit of the vehicle to a parameterized rounded rectangle given by: wherein the vehicle lies along the x and y axes with parameters including center coordinates xc, yc, major and minor axes a,b, an orientation Θ and exponent n; and passing parameters xc, yc, a, b, θ to a missile tracker. 18. The method of claim 17, wherein exponent n lies between 2 and 3 inclusive. 19. The method of claim 18, wherein exponent n lies between 2.1 and 2.6 inclusive. 20. The method of claim 17, wherein the Hough transform is computed as a sequence of 2D Hough transforms by fixing three of the parameters (xc, yc, a, b, θ), computing a 2D Hough transform for the other two parameters, and iterating on the three fixed parameters. 21. The method of claim 20, wherein the Hough transform is normalized by the perimeter of the parameterized object to determine the best fit. 22. The method of claim 17, wherein the image is acquired with a track gate, further comprising: Adjusting the position and size of the track gate based on said position, size and orientation parameters so that said track gate is approximately conformal with the vehicle. 23. An apparatus, comprising: a missile; an imager for acquiring an image of a scene containing a vehicle; an automatic target recognizer (ATR) for placing an initial non-conforming track gate over the vehicle in the image; and a missile tracker including a target acquisition module for acquiring a vehicle said target acquisition module performing edge detection on a portion of the image within the track gate to produce a silhouette image and performing a Hough transform on the silhouette image to fit a parameterized object to the vehicle and using the position, size and orientation parameters of the object to adjust the track gate so that it is approximately conformal with the vehicle, said missile tracker using the conformal track gate to track the acquired vehicle to impact. 24. The apparatus of claim 23, wherein the Hough transform is performed with parameters including center coordinates xc, yc, major and minor axes at a,b, and orientation Θ. 25. The apparatus of claim 23, wherein the imager acquires images of the vehicle through the conformal track gate, said target acquisition module periodically recomputing the Hough transform to update the parameters to maintain a track gate that is approximately conformal with the vehicle. 26. The apparatus of claim 23, wherein the missile tracker's targets acquisition module selects an aimpoint on the vehicle based on the position, size and orientation parameters. 27. An apparatus, comprising: A missile; a high resolution off-missile imager that acquires high resolution images of the scene; an off-missile target acquisition module that computes the Hough transform on the high resolution image and outputs the parameters; a low bandwidth channel that communicates the parameters to the missile; and a missile seeker including a low resolution on-missile imager that acquires low resolution images of the scene with a different line-of-sight than the high resolution imager; and a missile tracker having an on-missile target acquisition module that receives the parameters and recomputes the Hough transform on the low resolution images in a constrained space based on the parameters to compensate for the different line-of-sight and passes the parameters to the missile tracker. 28. A method of tracking a target vehicle from a missile, comprising: acquiring an image of a scene containing a vehicle; placing an initial track gate around the vehicle; using a missile tracker's target acquisition module to (a) perform edge detection on the portion of the image within the track gate to produce a silhouette image, (b) perform a Hough transform on the silhouette image to fit a parameterized object to the vehicle to extract position, size and orientation parameters of the object and (c) adjust the track gate based on said position, size and orientation parameters so that said track gate is approximately conformal with the vehicle; and passing the conformal track gate to the missile tracker to track the acquired vehicle to impact. 29. The method of claim 28, wherein an automatic target recognizer (ATR) places the initial track gate around the vehicle.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (5)
Choate William C. (Dallas TX), Method and apparatus for detecting edge spectral features.
Dennison, John Charles; Davis, Clayton Houston; Frazier, Brian Edward, Determining angular rate for line-of-sight to a moving object, with a body-fixed imaging sensor.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.