Model classification method based on frequency modulation continuous wave radar

[0031] The technical solutions of the present invention will be described in detail below with reference to the accompanying drawings.

[0032] like figure 1 As shown in the flowchart of the vehicle classification method based on FM continuous wave radar according to a specific embodiment of the present invention, firstly, the constant false alarm rate processing is performed to estimate the power spectrum of noise and interference and set the threshold, and then the power spectrum of the vehicle is estimated. Peak statistics. After the vehicle's power spectrum peaks are accumulated to a certain number, a histogram of the results is performed. Finally, according to the results of the histogram statistics, the power spectrum peak energy segmentation thresholds corresponding to different vehicle models are adaptively obtained, and finally the purpose of traffic statistics and vehicle classification is achieved.

[0033] like figure 2 According to a specific embodiment of the present invention, a schematic diagram of traffic radar installation and road environment is shown in the figure. On the basis of the built hardware platform 201, we conducted a field test on the flyover of the Jiangning Campus of Nanjing University of Aeronautics and Astronautics. The frequency-modulated continuous wave radar 201 is installed on the railing 202 of the overpass, and ensures that the beam direction of the radar is perpendicular to the traveling direction of the vehicle on the highway. Without loss of generality, we stipulate that the detection lane is selected as the second lane 203 .

[0034] like image 3 According to the result of the non-target constant false alarm rate processing in a specific embodiment of the present invention, the protection unit length of the constant false alarm rate processing is set to be L=50, the reference unit length is M=20, and the false alarm constant K 0 = 1.2. At this time, the probability of false alarm is P fa ≈1.4×10 -7. Due to the influence of the road surface on the radar echo signal and the influence of the noise of the radar system itself, it is shown in the form of the clutter power spectrum represented by the solid line. The dotted line part is the threshold value calculated by the constant false alarm rate. It can be seen that the threshold calculated by the constant false alarm rate can filter out most of the clutter and noise interference.

[0035] like Figure 4 According to the processing result of the target constant false alarm rate in a specific embodiment of the present invention, the power spectrum peak 401 corresponding to the vehicle echo is higher than the constant false alarm rate threshold 402 corresponding to the corresponding frequency point, so the target will not be affected by Filtered out by the constant false alarm rate threshold.

[0036] like Figure 5 According to a specific embodiment of the present invention, the statistics of the peak value of the power spectrum refers to the case where the peak value of the power spectrum when 100 vehicles pass by is recorded, that is, T=100.

[0037] like Image 6 According to the histogramization of the spectrum peak statistical array in a specific embodiment of the present invention, because the radar scattering cross-sectional areas of different vehicle types are different, the power spectrum peaks thereof are also different. The specific performance is that when the distance and angle between the radar and the target are the same, the radar scattering cross-sectional area of ​​the cart is large, and the corresponding power spectrum peak is also large. After performing histogram operation on the power spectrum peak statistical array, set H=40, three energy concentration regions 601, 602 and 603 are formed as a result. These three regions naturally form the energy distribution characteristics of the three models. Take the power spectrum peak value of 12.9 in the middle of 601 and 602 as the judgment threshold of Model 1, and the power spectrum peak value of 21.4 in the middle of 602 and 603 as the judgment threshold of Model 2. By increasing the level of the histogram, the types of identifiable vehicle models can be increased.

[0038] According to actual observations, during the statistical process, 1 container-type vehicle, 9 bus-type vehicles, and 90 small and medium-sized vehicles actually passed through the radar observation area. and Image 6 The classification results shown are consistent. Thus, the feasibility and accuracy of the method of the present invention for classifying vehicle types are verified.