Table 4 The accuracy declines (%) of different approaches as the number of images in each category changes from 5 to 200.
From: A Parallel Adaboost-Backpropagation Neural Network for Massive Image Dataset Classification
Classification approaches | 500 images | 1,000 images | 2,000 images | 5,000 images | 10,000 images | 20,000 images |
|---|---|---|---|---|---|---|
RF | 9.1 | 11.4 | 15.0 | 17.0 | 19.7 | 21.9 |
SVM | 9.2 | 10.8 | 14.4 | 16.8 | 18.4 | 20.1 |
Adaboost-BP | 9.1 | 9.9 | 13.1 | 15.1 | 17.7 | 20.7 |
The method of Shi et al. | 6.0 | 5.8 | 5.9 | 8.8 | 16.8 | 20.0 |
Parallel BP | 1.5 | 2.5 | 5.5 | 8.5 | 12.2 | 16.4 |
Parallel Adaboost-BP | 0.5 | 1.5 | 1.8 | 6.4 | 7.3 | 11.0 |
