matlab卷积逆投影,中国科学院沈阳自动化研究所

其他摘要The invention relates to a matrix reconstruction imaging method based on convolution adverse projection. The method comprises the following steps: establishing a radar coordinate system, determining an imaging region, and partitioning imaging units, wherein a MIMO radar array emits an orthogonal frequency-division signal and receives an echo signal; performing target detection on the imaging region and mapping to the imaging unit; extracting pixel units according to the probability so as to form an image pixel unit set; performing order imaging on the sampling set pixel unit according to the convolution adverse projection unit, and reconstructing other pixel units according to a matrix filling algorithm. By using the structure and relevancy feature of the two-dimension image data, the effective imaging pixel points and the auxiliary pixel points are extracted by using the priori information to realize the real-time online full-region image reconstruction; the operation of orderly imaging all pixel points in the target region is avoided, the operand is reduced, and the instantaneity requirement of the imaging algorithm is guaranteed.