This paper proposes a covariance based multiple watermark embedding using modified normalized principal components watermarking. Multiple watermarking increases the reliability of watermarks against various attacks. Blocks of the host image are selected for multiple watermark embedding by evaluating covariance between the blocks and textures of the watermark. Once the blocks and textures are identified, modified normalized principal components watermarking is applied in DWT-SVD domain to avoid false positive error. Performance of the proposed watermarking technique is analyzed by peak signal to noise ratio, correlation coefficient and structural similarity index. Comparative analysis with other watermarking methods also reveals the better performance of the proposed method. © 2018 IEEE.