Abstract: This study proposes a novel data-hiding method based on an improved neighbor mean interpolation and random-block division. The basic idea of the proposed method is to use the proposed neighbor mean interpolation as the scaling-up strategy and to make full use of the correlation between neighboring pixels to increase the payload without sacrificing much distortion for data hiding. Moreover, the random-block division technique is adopted to increase the security of data hiding method under the detection of a statistical steganalysis tool. Experimental results showed that our proposed method can obtain a large embedding capacity while maintaining high visual quality image and undetectability. Compared with another similar work, the PSNR is guaranteed to be higher than 34 dB which maintained a maximum capacity increase of 22.5-36.5%. Also, the undetectabilty can be improved by a maximum of 33.66%.