人造目标散射精细极化分解及特征提取方法研究

极化雷达目标散射精细刻画是微波遥感领域目标检测与识别中的紧迫课题和任务，但当今复杂电磁环境中异质散射、复杂结构、干扰混淆等现象的存在给其带来了极大的阻碍。传统目标极化分解由于数学规划狭隘，大都存在严重的电磁散射机理混淆问题。本项目立足于目标物理几何属性本质，通过优化和拓展物理散射模型，提出精细极化分解理念，更为全面、准确地揭示人造目标局部结构散射机制，并定量描述各散射机制主导性；在此基础上，结合精细极化分解获得的目标局部极化散射信息及散射功率，构造并提取多层次、多维度的目标极化特征，从散射精细刻画层面凸显目标与干扰间的细微差异，实现目标与干扰的准确检测与识别。项目的研究成果不仅可以丰富复杂电磁环境下极化雷达信息处理和目标解译理论，还可以增强极化雷达的抗干扰能力，推动极化雷达高精度、智能化目标检测与识别等领域的应用化进程。

Refined description of polarimetric radar target scattering is the urgent issue and task of target detection and recognition in the field of microwave remote sensing. However, it faces serious challenges posed by the existences of heterogeneous scattering, complex structures, and jammings in the complex electromagnetic environment. Due to the deficiency of mathematical programming, there exist severe electromagnetic scattering ambiguities in traditional polarimetric target decompositions. Based on the physical geometric properties, this project put forwards the concept of refined polarimetric target decomposition through optimizing and extending the physical scattering models. Accordingly, the scattering mechanisms of local structures of man-made targets are characterized, and the dominances of different scattering mechanisms are quantitatively described. By integrating the scattering information and scattering contributions derived from the refined polarimetric target decomposition, several polarimetric features are designed and extracted in a multilevel, multidimensional manner, which highlight the subtle differences between targets and jammings, thus the accurate detection and recognition is realized. On the one hand, the proposed concepts and methodologies are expected to enrich the theory of polarimetric radar information processing and target interpretation in the case of complex electromagnetic environment. On the other hand, the research findings can enhance the anti-jamming capability of polarimetric radar, thus further promoting the application process of high-precision and intelligentized target detection and recognition.