【作者】 秦小琦； 林家骏； 戴蒙；

【Author】 Qin Xiaoqi Lin Jiajun Dai Meng(School of Information Science and Engineering, East China University of Science and Technology, Shanghai 200237,China)

【机构】 华东理工大学信息科学与工程学院；

【摘要】 为了评价信息隐藏技术,建立正确、全面的不可感知性指标,在研究主观评价及人眼视觉特性基础上,分析传统图像质量评价指标峰值信噪比(PSNR)的缺陷和不足,并首次引入构造相似性函数(SSF)作为不可感知性评价指标.实验结果表明,该指标在频率特性、色感特性、方向特性上皆能正确评价隐写后图像视觉效果,其精确性、单调性优于现有指标.采用SSF函数作为信息隐藏中不可感知性指标,能弥补PSNR评价方法的局限性,更符合人眼主观视觉特性,具有实际应用价值.更多还原

【Abstract】 The evaluation of information hiding should be based on an accurate and complete imperceptibility index. Based on subjective image quality methods and human visual system (HVS), the structural similarity function (SSF) is first introduced into the practical steganography evaluation to balance the fault and limitation of traditional objective methods—peak signal to noise ratio (PSNR). The experiments prove that SSF, which can benchmark the visual effect of stego-images correctly in the frequent, color and direction properties, exceeds PSNR in precision and monotonicity. SSF, as the imperceptibility index, is better identical with the perceptual assessment of human beings and more practical than PSNR.更多还原