Research Article

[Retracted] Identifying Animals in Camera Trap Images via Neural Architecture Search

Table 3

Accuracy comparison based on MCTI.

ClassNASRT (ours)Resnet-18 [46]DenseNet [47]MobileNet-v2 [48]SGAS [44]SETN [26]PDARTS [24]

Agouti97.6095.2196.4195.8196.4189.8292.81
Bird98.6197.9297.9299.31100.0098.61100.00
Coiban agouti100.00100.00100.00100.0099.6198.06100.00
Collared peccary98.2596.4994.1592.9897.6692.4098.83
Opossum98.1594.4496.9196.3098.1593.2197.53
European hare97.4889.9296.6496.6496.6499.1694.96
Great tinamou99.5699.1198.2299.5699.5693.3397.78
Mouflon99.7799.3099.3099.0799.3097.9198.84
Ocelot89.4782.1185.2675.7982.1168.4274.74
Paca100.0092.0996.7495.3595.8190.7096.28
Red brocket deer96.5993.7596.0293.1891.4893.1892.61
Red deer100.00100.0099.80100.00100.0099.61100.00
Red fox95.6093.4192.3191.2193.4179.1293.41
Red squirrel97.3593.8194.6993.8194.6988.5087.61
Roe deer98.7298.2998.2998.2998.2996.5898.29
Spiny rat99.24100.0096.2191.6797.7384.8597.73
White-nosed coati99.5799.1598.7297.0297.8795.3297.45
White-tailed deer99.7598.9997.9898.2499.5095.9799.75
Wild boar99.7198.8299.4199.7199.4199.4199.71
Wood mouse100.00100.00100.00100.00100.00100.00100.00
Average98.2796.1496.7595.7096.8892.7195.92