IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
UP-0118079
(2008-05-09)
|
등록번호 |
US-7671786
(2010-04-21)
|
발명자
/ 주소 |
- Jain, Ankur
- Nikovski, Daniel N.
|
출원인 / 주소 |
- Mitsubishi Electric Research Laboratories, Inc.
|
대리인 / 주소 |
|
인용정보 |
피인용 횟수 :
1 인용 특허 :
7 |
초록
▼
A moving object is classified by transmitting, by a linear array of transmit antenna elements, a microwave into a surveillance area. A scattered microwave backprojected from a moving object is received by a linear array of receive antenna elements. Features are extracted from the scattered microwave
A moving object is classified by transmitting, by a linear array of transmit antenna elements, a microwave into a surveillance area. A scattered microwave backprojected from a moving object is received by a linear array of receive antenna elements. Features are extracted from the scattered microwave related to a spiral evolution of the scattered microwave. The moving object is then classified as one of a set of possible classes according to the extracted features, and an alarm signal can be generated indicating the selected class.
대표청구항
▼
We claim: 1. A method for classifying a moving object using microwaves, comprising: transmitting, by a linear array of transmit antenna elements, a microwave into a surveillance area; receiving, by a linear array of receive antenna elements, a scattered microwave backprojected from a moving object;
We claim: 1. A method for classifying a moving object using microwaves, comprising: transmitting, by a linear array of transmit antenna elements, a microwave into a surveillance area; receiving, by a linear array of receive antenna elements, a scattered microwave backprojected from a moving object; extracting features from the scattered microwave related to a spiral evolution of the scattered microwave, in which the features are extracted using a sliding time-window, in which the sliding window includes N samples x1, x2, . . . , xN, where x1 is a first sample, and XN is a last sample for any instance in time, and in which the features are a total curvilinear distance traversed as a result of the scattered microwave, expressed as CURVD = ∑ i = 2 N ( x i - 1 - x i ) 2 ; classifying the moving object as one selected class of a set of possible classes according to the extracted features; and generating an alarm signal indicative of the one selected class. 2. The method of claim 1, in which the possible classes includes people, vehicles and animals. 3. The method of claim 1, in which the linear array of transmit antenna elements and the linear array of receive antenna elements are in a form of leaky coaxial cables having slots punched into an outer conductor sheath, and further comprising: arranging the linear array of transmit antenna elements and the linear array of receive antenna elements at a perimeter of the surveillance area. 4. The method of claim 1, further comprising: classifying multiple different objects concurrently. 5. The method of claim 1, in which the features measure a displacement of the scattered microwave. 6. The method of claim 1, in which the moving object causes a complex signal received by each element of the linear array of receive antennas elements to rotate and translate generating the spiral evolution. 7. The method of claim 1, in which the sliding time-window is fixed. 8. The method of claim 1, in which the sliding time-window is variable. 9. The method of claim 1, in a new curvilinear distance CURVDnew based on an old curvilinear distance CURVDold is CURVDnew=CURVDold+(xN−xnew)2−(x1−x2)2. 10. The method of claim 1, in which the features are eigen-values obtained by a linear eigen-value analysis. 11. The method of claim 1, in which the classifying uses a support vector machine. 12. The method of claim 1, in which the classifying uses a k-nearest neighbor classifier. 13. The method of claim 1, in which the classifying uses a naïve Bayes classifier. 14. The method of claim 1, in which the classifying uses a sorted signal naïve Bayes classifier. 15. The method of claim 14, further comprising: sorting input samples of the scattered microwave according to signal values before extracting the features. 16. An apparatus for classifying a moving object using microwaves, comprising: a linear array of transmit antenna elements configured to transmit a microwave into a surveillance area; a linear array of receive transmit antenna elements configured to acquire a scattered microwave backprojected from a moving object; microwave related to a spiral evolution of the scattered microwave; means for extracting features from the scattered microwave related to a spiral evolution of the scattered microwave, in which the features are extracted using a sliding time-window, in which the sliding window includes N samples x1, x2, . . . , xN, where x1 is a first sample, and XN is a last sample for any instance in time, and in which the features are a total curvilinear distance traversed as a result of the scattered microwave, expressed as CURVD = ∑ i = 2 N ( x i - 1 - x i ) 2 . a classifier configured to classify the moving object as one selected class of a set of possible classes according to the extracted features; and an alarm signal indicating the one selected class.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.