Estimating a covariance matrix plays an essential role in modern statistics and practical applications. One key challenge is seeking an outperformed covariance matrix estimator in large dimension and small sample size scenarios. In this paper, we develop new covariance matrix estimators based on shrinkage regularization acclaimed in high-dimensional statistics. Individually, we consider two kinds of Toeplitz-structured target matrices in the shrinkage estimation. Firstly, we obtain the optimal tuning parameter under the mean squared error criterion in closed form by discovering the mathematical properties of the two target matrices. Secondly, we get some vital moment properties of the complex Wishart distribution, and then we simplify the optimal tuning parameter as the data come from the complex Gaussian distribution. Thirdly, we unbiasedly estimate the unknown scalars involved in the optimal tuning parameter. Then we immediately propose available shrinkage estimators. To verify the performance of the proposed covariance matrix estimators, we provide some numerical simulations and applications to array signal processing compared to some existing estimators.