EURASIP Journal on Advances in Signal Processing

Table 8 Accuracy (%) of seal calls recognition by the RBM [10] model

From: Seal call recognition based on general regression neural network using Mel-frequency cepstrum coefficient features

SNR/dB	MFCC				LOFAR spectrum
SNR/dB	A	B	C	Total	A	B	C	Total
20	94.06	93.68	91.83	93.19	90.36	87.38	85.69	87.81
15	92.42	93.08	90.65	92.05	87.80	86.71	83.25	85.92
10	91.02	86.23	87.32	88.19	82.53	79.50	79.62	80.55
5	82.38	79.43	74.95	78.92	77.08	74.29	73.72	75.03
0	75.97	69.27	65.03	70.09	64.05	65.94	60.57	63.52

Back to article page