In this paper, a secure watermarking algorithm is proposed to efficiently enhance invisibility and robustness simultaneously. First, the combination of transform domain scheme is used to identify the appropriate coefficient for data embedding. To offer a good balance between robustness and invisibility, we computed optimal embedding factors by using the fusion of particle swarm optimisation (PSO) and the Firefly algorithm. Subsequently, the watermark is scrambled by using a step space-filling curve, and the scrambled watermark is embedded into the wavelet domain of the cover image via the optimal embedding factor. Furthermore, selective encryption is applied to the original host image so that the proposed algorithm can also provide an additional level of security at low cost. We achieved average scores of peak signal-to-noise ratio (PSNR), normalised coefficient (NC), number of changing pixel rate (NPCR) and unified averaged changed intensity (UACI) as 55.1471 dB, 0.9956, 0.9801 and 0.3471, respectively.

在本文中，提出了一种安全水印算法，以同时有效地增强不可见性和鲁棒性。首先，使用变换域方案的组合来识别数据嵌入的合适系数。为了在鲁棒性和不可见性之间取得良好的平衡，我们通过使用粒子群优化 (PSO) 和 Firefly 算法的融合来计算最佳嵌入因子。随后，利用阶梯空间填充曲线对水印进行加扰，并通过最优嵌入因子将加扰后的水印嵌入到覆盖图像的小波域中。此外，选择性加密应用于原始主机图像，因此所提出的算法还可以以低成本提供额外的安全级别。我们达到了峰值信噪比 (PSNR) 的平均分数，