Computational Intelligence and Neuroscience / 2019 / Article / Tab 1

Research Article

An Improved Convolutional Neural Network Algorithm and Its Application in Multilabel Image Labeling

Table 1

Comparison of the accuracy of labeling for each category in the Pascal VOC 2007 and Pascal VOC 2012 datasets based on different algorithms.

Image categoryLabeling accuracy
Zuo et al. [26]Zhang et al. [27]Islam et al. [28]CNNDCCNN
2007201220072012200720122007201220072012

Plane0.8020.7770.9880.9730.9470.9240.9920.9831.00.999
Bike0.5010.4250.8120.7480.4980.4510.9050.8770.9960.973
Bird0.5610.4540.8730.8080.9620.9461.00.9771.00.984
Boat0.6190.5330.8990.8530.6710.6520.9350.9200.9900.972
Bottle0.280.240.6910.6080.7910.7580.8950.8790.9240.919
Bus0.7840.7220.9310.8990.9660.9510.9760.9710.9860.980
Car0.5840.5060.8970.8680.9050.8910.9530.9490.9890.987
Cat0.6070.5420.9410.8930.9410.9230.9620.9550.9840.970
Chair0.5090.4530.6130.5540.4220.390.8260.7940.9070.893
Cow0.3090.260.8480.7780.8660.8571.00.9991.01.0
Dining table0.3980.3660.8290.7510.7490.7040.8430.8250.8990.885
Dog0.5070.4260.8850.830.8950.8860.9180.9050.9920.971
Horse0.4410.3890.9160.8750.9120.8940.9290.9270.9800.978
Motorbike0.570.5070.8250.7920.7980.7610.8510.8490.9310.931
Person0.7690.7030.8990.8470.8310.7940.8990.8970.9620.957
Potted plant0.3050.2340.6360.5780.6810.6580.8330.8290.8840.881
Sheep0.4060.3620.8240.7920.8950.8620.9600.9600.9960.993
Sofa0.3850.3140.4490.3950.3890.3390.7250.7190.8190.816
Train0.6990.6160.9350.9060.8640.8181.00.9891.01.0
TV monitor0.5230.430.7930.7740.7850.7290.8190.8160.8570.856
MAP value0.5280.4630.8240.7760.7880.7590.9110.9010.9550.947