Table 3 Regression results between seedling counts obtained by the proposed method and human field assessment
Detection model | Testing videos | Regression equation | Video quantity | \(R^{2}\) | RMSE | MAE | MRE (%) |
|---|---|---|---|---|---|---|---|
\(model_{SAll}\) | All | Y = X | 75 | 0.98 | 1.6 | 1.5 | 7 |
\(model_{SAll}\) | TAMU2015 | Y = 0.96 X | 25 | 0.85 | 3.3 | 3.4 | 11 |
\(model_{SAll}\) | UGA2018 | Y = X | 25 | 0.99 | 0.2 | 0.5 | 6 |
\(model_{SAll}\) | UGA2015 | Y = 1.01 X | 25 | 0.96 | 1.1 | 0.8 | 4 |
\(model_{U15U18}\) | TAMU2015 | Y = 0.95X | 25 | 0.80 | 3.9 | 3.4 | 11 |
\(model_{U15T15}\) | UGA2018 | Y = 1.02X | 25 | 0.68 | 0.8 | 0.6 | 7 |
\(model_{T15U18}\) | UGA2015 | Y = 0.32X + 18.65 | 25 | 0.33 | 4.3 | 12.1 | 57 |