基于JPEG_LS的图像无损信息隐藏方法

Lossless Information Hiding Algorithm in Image Based on JPEG_LS Prediction

使用JPEG_LS模板对宿主图像进行预测.信息嵌入时,直接使用t位信息替换预测残差的后t位,以减小预测残差改变及嵌入后的图像失真.分析算法中所用阈值及嵌入容量对图像峰值信噪比的影响.试验表明:改进算法在与原算法嵌入容量相同的情况下,峰值信噪比提高了十几d B,并且可以做到图像的无损恢复.

JPEG_LS algorithm is applied to predict the original host image.When information is embedded,the last t bits of prediction-error are replaced by the last t bits of embedding information directly to decrease the change of prediction-error and the distortion of embedded image.The influence on the peak signal to noise ration caused by thresholds and capacity in the algorithm is also analyzed.Peak signal to noise ration of the proposed algorithm is increased by more than 10dB in comparison with the traditional way at the same case of embedded capacity,and the embedded image could still be retrieved after extracting hidden data.