Skip to main content

Table 6 Summary of data annotation and partitioning for the combined datasets

From: DeepSeedling: deep convolutional network and Kalman filter for plant seedling detection and counting in the field

Dataset

T15U15

T15U18

U15U18

\(Seedling_{All}\)

Total number of images

4099

3715

3406

5610

Number of training images

3404

3054

2856

4657

Number of validation images

695

661

550

477

Number of testing images

N/A

N/A

N/A

476

Total number of annotations (plant seedling/weed)

29739 (27997/1742)

27809 (27025/794)

13696 (12736/960)

35597 (33862/1735)

Number of training annotations (plant seedling/weed)

24466 (23035/1431)

22805 (22154/661)

11388 (10606/782)

29325 (27895/1430)

Number of validation annotations (plant seedling/weed)

5273 (4962/311)

5004 (4871/133)

2308 (2130/178)

3147 (3004/143)

Number of testing annotations (plant seedling/weed)

N/A

N/A

N/A

3125 (2963/162)