Table 2.
Statistics of field phenotype research on dicotyledonous crops.
| Classification of indicators | Crop category | Type of data | Phenotypic analysis method | Phenotypic parameters | Accuracy % | R 2 | Shooting scale | Year | Author |
|---|---|---|---|---|---|---|---|---|---|
| Morphological indicators | Cotton | RGB | 3D reconstruction | Stem height, leaf width, leaf length | 91.66, 94.25, 91.22 | – | Single | 2012 | Paproki et al., 2012 |
| Soybean | Thermal, Multispectral | Machine learning | Canopy coverage, canopy height | – | 0.86, 0.99 | Single | 2016 | Kirchgessner et al., 2016 | |
| Cotton | RGB | CNN | Number of flowers | Error = −4~3 | – | Single | 2017 | Xu et al., 2017 | |
| Rapeseed | Multispectral, RGB | Machine learning | Canopy coverage | – | 0.79 | Group | 2021 | Wan et al., 2021 | |
| Soybean | RGB | Machine learning | Canopy coverage, canopy height | 90.4, 99.4 | – | Group | 2020 | Borra-Serrano et al., 2020 | |
| Cotton | RGB | CNN | Flowering patterns | – | 0.88 | Single | Jiang et al., 2020 | ||
| Soybean | RGB | SFM | Canopy roughness | – | >0.5 | Group | Herrero-Huerta et al., 2020 | ||
| Cotton | RGB | Metashape, Python | Canopy coverage | 93.4 | – | Group | 2021 | Xu et al., 2021 | |
| Arabidopsis | RGB | CNN | Number of leaves | – | 0.92 | 2020 | Dobrescu et al., 2020 | ||
| Cotton | Lidar | 3D point cloud | Plant height | – | 1 | Single | 2017 | Sun et al., 2017 | |
| Physiological and biochemical indicators | Soybean | RGB | Machine learning | Leaf iron deficiency chlorosis | >81, 96 | – | Regional | 2018 | Bai et al., 2018 |
| 2017 | Naik et al., 2017 | ||||||||
| Cotton | Near Infrared Spectroscopy | / | Leaf macro and micronutrients | 87.3, 86.6 | – | Organ | 2021 | Prananto et al., 2021 | |
| Soybean | Hyperspectral | DNN | Fresh biomass of above ground | – | 0.91 | Group | 2021 | Yoosefzadeh-Najafabadi et al., 2021 | |
| Cotton | Hyperspectral | / | Coverage, water use efficiency | – | – | Group | 2018 | Thorp et al., 2018 | |
| Soybean | Spectral Scanner | Modeling | εe, εc | – | 0.68 | Organ | 2021 | Keller et al., 2021 | |
| Biotic/Abiotic Stress | Rapeseed | RGB | CNN | Oilseed rape pests | 77.14 | Regional | 2019 | He et al., 2019 | |
| Rapeseed | RGB | Machine learning | Fruiting bodies of Leptosphaeria maculans | – | 0.87 | Regional | 2019 | Bousset et al., 2019 | |
| Soybean | RGB | DCNN | Nonbiological | – | – | Regional | 2018 | Ghosal et al., 2018 | |
| Soybean | RGB | Machine learning | Leaf iron deficiency chlorosis | 96% | – | Single | 2018 | Naik et al., 2017 | |
| Soybean | Multispectral, Infrared |
Machine learning | Flood | – | 0.9 | Organ | 2021 | Zhou et al., 2021 | |
| Yield | Soybean | RGB | / | Canopy coverage | – | 0.4–0.7 | Regional | 2016 | Bai et al., 2016 |
| Soybean | RGB | Machine learning | Yield and maturity | – | 0.51, 0.82 | Group | 2020 | Borra-Serrano et al., 2020 | |
| Soybean | RGB | / | Yield/canopy cover | – | 0.75 | Group | 2019 | Moreira et al., 2019 | |
| Soybean | Hyperspectral | DNN (EB) | Yield | – | 0.76, 0.77 | Group | 2021 | Yoosefzadeh-Najafabadi et al., 2021 |
RGB, an abbreviation for the three primary colors; CNN, convolutional neural network; DCNN, deep convolutional neural network; SFM, multiview structure from motion; EB, integrated baggies εe, photochemical energy; εc, biomass.
The slash (/) indicates the ratio.
The short line (–) indicates that it is not mentioned in the article.