Fig. 1. The signal in ideal or the baseline having linear distortion and spatial representation of saliency
(a): Ideal positive signal; (b): Ideal negative signal;(c): Positive signal with linear distortion in the baseline; (d): Negative signal with linear distortion in the baseline

Fig. 2. The signal of baseline having linear distortion and spatial representation of saliency
(a): Positive signal and its saliency space; (b): Negative signal and its saliency space

Fig. 3. The process of signal detection

Fig. 4. Signal extraction with noise and the baseline in nonlinear distortion

Fig. 5. Results of signal extraction from 100 experiments with different signal-to-noise ratios, different baselines, and different signals
(a): The mean value of the absolute error of the signal extraction result;(b): The root mean square value of the absolute error of the signal extraction result;(c): The mean value of the root mean square error of the signal extraction results;(d): The mean square root of the mean square error of the signal extraction results

Fig. 6. Comparison of extraction effects of different extraction algorithms

Fig. 7. The extraction results of different algorithms to different signals under different signal-to-noise ratios and different baseline types
(a): The mean value of absolute error of signal 1 extraction result;(b): The root mean square value of the absolute error of the signal 1 extraction result;(c): The mean value of absolute error of signal 2 extraction result;(d): The root mean square value of the absolute error of the signal 2 extraction result

Table 1. Comprehensive extraction effect of different algorithms