Two-stage CFAR target detection method based on weighted fusion filtering

In complex scenes, such as leaf vegetation occlusion and strong reflection from thick branches, the signal-to-noise ratio (SNR) of the target echo signal received by the UAV radar is low, and its synthetic aperture radar (SAR) image shows obvious pseudo-target bright spots, which makes it difficult to detect interesting targets based on the image domain and has a high false alarm rate. To solve the above problems, the prerequisite is to filter the SAR image for preprocessing. However, different filters have inconsistent filtering effects on SAR images in different environments and are not universal. Based on the principle of adaptive weighted fusion filtering, a two-stage constant false alarm rate (TSCFAR) detection method is proposed in this work, which comprehensively utilizes the filtering effects of different filters to remove strong interference clutter from SAR images. By retaining more detailed features, a high probability of target detection is achieved. Finally, the effectiveness of the algorithm is verified using measured data.