In this study, a low-complexity watermarking for embedding watermark into two or more halftone images with dot diffusion is proposed. The first halftone image is obtained by traditional dot diffusion, and the others are achieved by applying the proposed Secret-Shared Dot Diffusion technique (SSDDF). The visual decoding pattern can be observed when these similar dot-diffused images are printed on transparencies and then overlaid each other. Better decoded result can be obtained by computer-aid XNOR operation. Compared to the Dot Diffusion with Nonlinear Thresholding (DDNT) proposed by Taheri et al, the correct decoding rate can be improved with the proposed approach. Moreover, two extensions called dual SSDDF (DSSDDF) and Adaptive SSDDF (ASSDDF) are proposed to provide better decoding rates and compromised image quality.