太赫兹ISAR增强成像方法研究OA

To improve the imaging quality of terahertz Inverse Synthetic Aperture Radar(ISAR)and address the limitations of existing adaptive aperture scaling transform algorithms,this paper proposes a maximum Carrier-to-Noise Ratio(CNR)scaling transform algorithm based on scatterer clustering.The method is founded on the analysis of regional statistical characteristics of the image:image patches are first classified according to background clutter statistics,followed by target scatterer segmentation.Hierarchical Density-Based Spatial Clustering of Applications with Noise(HDBSCAN)is then employed to cluster the scatterers.Finally,an optimization problem is solved to maximize the CNR of target scatterer clusters within each sub-aperture.Experimental results validate the effectiveness of the proposed approach,demonstrating significant CNR improvement in the imaging results,clearer target contours,and enhanced weak scatterers.Compared with existing algorithms,this method innovatively introduces a maximum CNR scaling transform algorithm that segments targets and clusters scatterers to partition scatterer clusters for each sub-aperture.The algorithm maximizes the CNR of the fused image target without altering the energy distribution of scatterers within clusters,thereby enhancing weak scatterer contours while preserving target details,and alleviating the blocking artifacts problem arising from insufficient enhancement of weak scattering regions in conventional methods.