Underwater multi-bright-point target detection method based on compression perception

The invention discloses an underwater multi-bright-point target detection method based on compression perception. The underwater multi-bright-point target detection method comprises the steps that: (1) emission baseband signals can form an over-complete atom database, and receiving signals represent products of the over-complete atom database and bright point structures; (2) a matched tracking algorithm of a compression perception technology is utilized to estimate the bright point structures of a target; (3) the bright point structures obtained by estimation are utilized to modify the coefficient of a matched filter; and (4) the modified coefficient of the matched filter is utilized to carry out matched filtering on a target echo. According to the invention, According to the invention, the output of a hedgehog-shaped matched filter is improved into brief single-peak output in the underwater multi-bright-point target detection process, so that the signal processing gain is improved, the false alarm number is