Aiming at solving the decline of detection performance under low signalnoise ratio and nonstationary background noise and combining the principle of eigenvalue detection, a combination information cross-correlation detection algorithm based on single vector hydrophone was presented. This algorithm makes an assemble velocity by using electronic rotation angle and velocity information, and obtains a cross-correlation value with pressure. This value satisfies the asymptotic Gaussian distribution under large snapshot without target signal. This value is divided by minimum eigenvalue of analytical velocity covariance matrix to get a detection statistic. Finally, compared with the threshold, the object detection is achieved. The analysis of theory shows this algorithm does not need to know any prior information of background noise, and the detection performance can be improved by adjusting the guiding angle. This algorithm can achieve bearing estimation by using the relationship between detection statistic and guiding angle as to single target. The simulation and real data prove the superiority of the proposed algorithm, compared with the maximum-minimum eigenvalue detection algorithm and the energy detection algorithm.