基于分段线性坐标优化的退化压缩感知水印检测

DETECTION OF DEGRADED COMPRESSED SENSING WATERMARK BASED ON PIECEWISE LINEAR COORDINATE OPTIMIZATION

为提高水印检测算法的性能,提出一种基于搜索优化改进坐标法的退化压缩感知水印检测算法。为提高水印检测算法模型的精度,基于退化压缩感知算法对扩频水印检测模型进行构建。以最小化失真(包括CS重建和水印检测)作为优化指标,采用检测概率为特征的检测可靠性约束,进行最优测量次数(提供带宽)和嵌入强度的优化。采用图论处理方式进行坐标搜索算法的优化过程改进,获得一组能够符合约束限制的解集,实现对水印检测算法性能的提升。通过仿真实验,对水印检测算法的性能进行了验证。

In order to improve the performance of watermark detection algorithm, we proposed degraded compressed sensing watermark detection algorithm based on search optimization and improved coordinate method. To improve the accuracy of the watermark detection algorithm model, the spread spectrum watermark detection model was constructed based on the degenerate compression sensing algorithm. Minimum distortion, including CS reconstruction and watermark embedding, was taken as the optimization index. We used the detection probability as the detection reliability constraint to optimize the measurement times （providing bandwidth） and embedded strength. The optimization process of the coordinate search algorithm was improved by using the graph theory processing method, and a set of solutions which met the constraints could be obtained to improve the performance of the watermark detection algorithm. The performance of the watermark detection algorithm is verified by simulation experiments.