In this paper, a new algorithm of multi-image watermarking is proposed for the security of multimedia data by using the features of Lifting wavelet transform (LWT), Singular value decomposition (SVD), Chirp Z-transform (CZT), and Arnold transform (AT). In the embedding procedure , split the input image into sub-bands using lifting wavelet transform and select HL and LH sub-bands for image watermarking, then apply the chirp Z-transform in the selected sub-bands and then decompose these two sub-bands by using SVD. Two watermark images are selected for embedding them into the input image, and the security of the second watermark is increased by using Arnold transform. Apply the lifting wavelet transform on both the watermark images and then transform them by SVD. Singular values (SVs) of watermarks are used to embed into SVs of the input image. The grey wolf optimization (GWO) technique is preferred here to calculate the scaling factor. Scaling factor plays very important role in the performance of watermarking. The effectiveness of the proposed algorithm is checked by applying different types of attacks. The superiority of the proposed work is mentioned in the results. Peak signal to noise ratio (PSNR), Normalized correlation coefficient (NCC) & Mean square error (MSE) is used for performance evaluation.