Skip to main content

Table 2 Mean absolute errors (MAE) and mean squared errors (MSE) for maize tassels counting on the test set of MTC dataset

From: Maize tassels detection: a benchmark of the state of the art

Method Sequences Overall
  Zhengzhou2011 Taian2010_2 Taian2011_2 Taian2012_2 Taian2013_2 Gucheng2014 Jalaid2015_2 Jalaid2015_3   
  MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE MAE MSE
JointSeg 20.9 23.2 46.6 47.9 16.4 19.7 25.1 29.8 6.5 8.0 7.3 10.5 27.8 29.1 53.2 61.3 24.2 31.6
mTASSEL 9.8 14.9 18.6 22.1 11.6 12.7 5.3 7.8 13.1 16.6 31.1 35.3 16.2 18.0 46.6 51.0 19.6 26.1
GlobalReg 19.0 21.5 23.0 24.7 14.1 16.8 13.5 15.7 19.6 25.2 19.5 21.7 11.2 13.7 42.1 45.4 19.7 23.3
DensityReg 16.1 20.2 9.9 10.7 9.2 11.7 10.8 12.7 20.2 23.7 9.4 10.5 7.2 7.9 23.5 26.9 11.9 14.8
CCNN 21.3 23.3 28.9 31.6 12.4 16.0 12.6 15.3 18.9 23.7 21.6 24.1 9.6 12.4 39.5 46.4 21.0 25.5
TasselNet 4.9 6.1 5.2 6.6 2.5 2.9 4.8 5.8 4.0 5.0 5.3 6.5 16.0 16.6 20.7 25.2 6.6 9.6
\(\hbox {TasselNet}^*\) 3.0 3.4 3.2 4.2 3.2 3.7 5.2 7.1 3.5 4.8 3.4 4.0 18.3 19.2 16.3 19.8 5.4 8.4
FaceBoxes 7.3 8.4 3.7 4.4 3.3 4.2 3.3 4.4 4.9 6.3 4.1 6.1 18.5 19.2 19.3 22.1 6.4 9.4
YOLOv3 4.3 5.7 5.0 6.6 2.9 3.9 3.0 4.0 4.5 6.0 2.3 2.8 20.3 21.7 24.1 25.9 6.1 9.9
RetinaNet 3.3 4.3 5.6 6.4 4.4 6.6 3.3 4.0 3.1 4.0 4.2 5.1 7.6 8.1 26.5 32.9 5.9 10.7
Faster R-CNN 2.6 3.1 3.2 4.0 2.9 3.4 2.6 3.2 3.6 4.4 7.6 9.0 8.6 9.6 21.3 25.5 5.8 8.9
  1. The two lowest error are shown in bold and bold italic