Skip to main content

Table 1 Overall performance of the model generalizability

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

Training setTesting setmAP \(\text {IOU}_{\text {all}}\)mAR100 \(\text {IOU}_{\text {all}}\)F1 \(\text {IOU}_{\text {all}}\)mAP \(\text {IOU}_{\text {0.5}}\)mAR100 \(\text {IOU}_{\text {0.5}}\)F1 \(\text {IOU}_{\text {0.5}}\)
T15U15 trainingUGA2018 testing0.7630.8080.7850.9811.0000.991
UGA2018 trainingUGA2018 testing0.7780.8180.7980.9830.9920.987
T15U18 trainingUGA2015 testing0.1790.3350.2330.3520.6830.464
UGA2015 trainingUGA2015 testing0.5990.6950.6430.9230.9970.959
U15U18 trainingTAMU2015 testing0.3770.5350.4420.5880.8570.698
TAMU2015 trainingTAMU2015 testing0.7910.8270.8090.9891.0000.995
  1. For the testing set in each of the UGA2018, UGA2015, and TAMU2015 datasets, two Faster RCNN models were trained using the training set from the same dataset and from the combination of the other two datasets, respectively. Performance comparison of the two models was used to evaluate the model generalizability