On the other hand, the TWD applied during the period of DWR imposition varied substantially between the WRs applied in the three cropping seasons.
In the period after DWR application, the ID applied in 2012-2013 was 330.1 mm, superior to those applied in 2011-2012 (184.3 mm) and 2013-2014 (171.6 mm), precisely to complement the greater ETo-Pef difference of 363.4 mm occurred in this period (Figure 1).
For all cultivars, the means of TCH in the cropping season of 2011-2012 were superior to those of 2012-2013 and 2013-2014, a consequence of the better water availability in the soil in this season, especially due to the occurrence of rainfalls that totaled 187.8 mm during the period of DWR imposition (Figure 1).
For a practical additive spread spectrum watermarking scenario, a is usually not determined arbitrarily, but it is based on the Data-to-Watermark (DWR) ratio, expressed in decibel (dB), where the term data refers to the coefficients used for embedding in one specific NSCT subband, which is determined by the energy distribution.
DWR =10 [log.sub.10] ([[sigma].sup.2.sub.x]/[[alpha].sup.2][[sigma].sup.2.sub.w], (29)
Hence, we can express the embedding strength a as a function of the DWR and the variance of the host signal as
[alpha] = square root of [[sigma].sup.2.sub.x]/exp ((log (10) * DWR)/10).
As we can see, with the increase of DWR, which means the decrease of the embedding strength, the probability of miss will increase, which is in accordance with theoretical analysis, since the smaller the embedding strength is, the much more difficult the watermark can be detected, and vice versa.
Caption: Figure 7: Performance evaluation of ROC (DWR = 6).
Caption: Figure 8: Performance evaluation of ROC (DWR = 8).
Caption: Figure 9: Performance evaluation of ROC (DWR = 10).