Skip to main content

Table 3 Comparison of the accuracy measure applied on the spectrogram

From: A novel approach to extracting useful information from noisy TFDs using 2D local entropy measures

SNRProposed methodLocal entropy-based algorithmRICI TFD threshold5%10%15%
Signal 1
−3 dBAcc= 0.983Acc= 0.964Acc= 0.982Acc= 0.887Acc= 0.959Acc= 0.969
0 dBAcc= 0.985Acc= 0.974Acc= 0.984Acc= 0.897Acc= 0.97Acc= 0.979
3 dBAcc= 0.993Acc= 0.978Acc= 0.992Acc= 0.988Acc= 0.985Acc= 0.978
6 dBAcc= 0.993Acc= 0.965Acc= 0.992Acc= 0.919Acc= 0.919Acc= 0.919
10 dBAcc= 0.994Acc= 0.986Acc= 0.972Acc= 0.919Acc= 0.919Acc= 0.919
Signal 2
−3 dBAcc= 0.961Acc= 0.967Acc= 0.953Acc= 0.868Acc= 0.933Acc= 0.942
0 dBAcc= 0.973Acc= 0.967Acc= 0.973Acc= 0.91Acc= 0.956Acc= 0.955
3 dBAcc= 0.983Acc= 0.95Acc= 0.981Acc= 0.972Acc= 0.961Acc= 0.948
6 dBAcc= 0.983Acc= 0.983Acc= 0.975Acc= 0.912Acc= 0.912Acc= 0.912
10 dBAcc= 0.987Acc= 0.982Acc= 0.983Acc= 0.912Acc= 0.912Acc= 0.912