Table 2.
Summary of boll count per tile in different counting ranges for each of the train, validation, and test datasets. Train and validation set contain tiles obtained from 285 full-scale raw images whereas test set represents tiles from 5 held-out images. Because of the dense boll population, the tiles containing more than 15 bolls cannot be counted accurately. Hence, these were not considered (DNC) in training or in the total count for the training and validation sets.
| Boll Count per Image | Training Tiles | Validation Tiles | Testing Tiles | Total |
|---|---|---|---|---|
| 0 | 919 | 50 | 21 | 990 |
| [1, 5] | 1852 | 102 | 48 | 2002 |
| [6, 10] | 710 | 42 | 6 | 758 |
| [11, 15] | 231 | 6 | 3 | 240 |
| above 15 | 300 (DNC) | 54 (DNC) | 6 | 360 |
| Total | 3712 | 200 | 84 | 4350 |