Journal of Spectroscopy / 2019 / Article / Tab 7 / Research Article
Identification of Tilletia foetida , Ustilago tritici , and Urocystis tritici Based on Near-Infrared Spectroscopy Table 7 Pathogen identification results of the BPNN models built with different modeling ratios in each modeling spectral region when Data treatment method 2 was used.
Spectral region (cm−1 ) Modeling ratio of training set to testing set The number of neurons in the hidden layer Training sets Testing sets Average identification accuracy for the 10 data subsets (%) Standard deviation of the identification accuracies for the 10 data subsets Average identification accuracy for the 10 data subsets (%) Standard deviation of the identification accuracies for the 10 data subsets 4000–12000 3 : 1 100 99.70 0.006994 99.78 0.006660 4000–12000 4 : 1 10 98.00 0.04237 97.22 0.04648 4000–12000 5 : 1 50 98.02 0.04058 97.67 0.05176 4000–5000 3 : 1 50 92.14 0.1541 92.00 0.1508 4000–5000 4 : 1 100 96.57 0.04367 96.39 0.04488 4000–5000 5 : 1 100 92.88 0.1175 93.33 0.1183 4000–6000 3 : 1 50 99.54 0.009770 99.56 0.01332 4000–6000 4 : 1 50 99.71 0.008580 99.44 0.01668 4000–6000 5 : 1 10 99.59 0.01024 99.33 0.01332 4000–7000 3 : 1 10 98.02 0.04279 97.78 0.04662 4000–7000 4 : 1 50 95.29 0.08590 95.00 0.08766 4000–7000 5 : 1 50 99.59 0.01024 99.67 0.009990 4000–8000 3 : 1 10 99.31 0.01617 99.33 0.01017 4000–8000 4 : 1 50 100 0 100 0 4000–8000 5 : 1 50 100 0 100 0 4000–9000 3 : 1 10 99.92 0.002280 99.78 0.006660 4000–9000 4 : 1 10 99.57 0.009156 99.17 0.01780 4000–9000 5 : 1 50 99.93 0.002040 99.67 0.009990 4000–10000 3 : 1 10 99.77 0.004893 99.78 0.006660 4000–10000 4 : 1 10 99.64 0.007314 99.44 0.01668 4000–10000 5 : 1 10 99.73 0.006261 100 0 5000–6000 3 : 1 50 99.54 0.01143 99.11 0.02038 5000–6000 4 : 1 50 95.14 0.1272 94.44 0.1400 5000–6000 5 : 1 50 100 0 100 0 5000–7000 3 : 1 10 99.62 0.009181 99.33 0.01421 5000–7000 4 : 1 10 99.29 0.01153 98.89 0.01844 5000–7000 5 : 1 10 97.67 0.05708 97.33 0.05925 5000–8000 3 : 1 10 99.85 0.004590 99.56 0.008880 5000–8000 4 : 1 50 99.86 0.004290 99.44 0.01112 5000–8000 5 : 1 10 99.80 0.006150 100 0 5000–9000 3 : 1 100 99.92 0.002280 99.78 0.006660 5000–9000 4 : 1 50 99.79 0.004573 99.44 0.01668 5000–9000 5 : 1 100 99.80 0.004381 99.67 0.009990 5000–10000 3 : 1 50 100 0 100 0 5000–10000 4 : 1 50 100 0 99.72 0.008340 5000–10000 5 : 1 50 99.66 0.01026 99.67 0.009990 6000–7000 3 : 1 50 94.28 0.1015 94.22 0.1010 6000–7000 4 : 1 50 95.72 0.08484 94.72 0.08094 6000–7000 5 : 1 50 88.77 0.1922 87.67 0.1700 6000–8000 3 : 1 50 100 0 99.78 0.006660 6000–8000 4 : 1 50 99.86 0.004290 99.44 0.01112 6000–8000 5 : 1 50 97.88 0.04379 97.00 0.06046 6000–9000 3 : 1 100 98.24 0.04069 98.00 0.04604 6000–9000 4 : 1 50 99.86 0.002840 99.44 0.01112 6000–9000 5 : 1 50 99.93 0.002040 99.67 0.009990 6000–10000 3 : 1 100 99.47 0.01186 99.33 0.01421 6000–10000 4 : 1 50 97.93 0.04233 96.67 0.04779 6000–10000 5 : 1 10 99.04 0.02435 99.33 0.01332 7000–8000 3 : 1 50 99.85 0.003040 99.33 0.01421 7000–8000 4 : 1 50 99.86 0.004290 99.44 0.01668 7000–8000 5 : 1 50 99.93 0.002040 99.00 0.02135 7000–9000 3 : 1 50 98.24 0.04040 97.33 0.04534 7000–9000 4 : 1 50 98.43 0.03805 97.22 0.04648 7000–9000 5 : 1 50 97.95 0.04029 97.33 0.05122 7000–10000 3 : 1 50 95.65 0.08542 95.56 0.08374 7000–10000 4 : 1 50 99.29 0.01356 98.33 0.02833 7000–10000 5 : 1 50 99.66 0.006296 99.33 0.01332 8000–9000 3 : 1 50 97.79 0.04002 97.11 0.03854 8000–9000 4 : 1 50 98.00 0.03949 96.39 0.05423 8000–9000 5 : 1 50 97.74 0.04030 96.67 0.04945 8000–10000 3 : 1 10 94.12 0.1173 93.56 0.1207 8000–10000 4 : 1 10 96.14 0.06422 95.28 0.05700 8000–10000 5 : 1 10 97.13 0.04351 96.67 0.05579 9000–10000 3 : 1 50 93.89 0.08533 93.56 0.08401 9000–10000 4 : 1 50 97.36 0.05689 96.11 0.07049 9000–10000 5 : 1 50 97.74 0.04076 97.00 0.05261