A Novel Hybrid GOA-XGB Model for Estimating Wheat Aboveground Biomass Using UAV-Based Multispectral Vegetation Indices
Abstract
:1. Introduction
2. Materials and Methods
2.1. Study Region
2.2. Data Collection
2.3. Artificial Intelligence Methods
2.3.1. Multilayer Perceptron (MLP)
2.3.2. Gaussian Process Regression (GPR)
2.3.3. Support Vector Machine (SVM)
2.3.4. Random Forest (RF)
2.3.5. Gradient Boosting Decision Tree (GBDT)
2.3.6. XGBoost
2.3.7. Grasshopper Optimization Algorithm (GOA)
2.3.8. Particle Swarm Optimization (PSO) Algorithm
2.3.9. Whale Optimization Algorithm (WOA)
- (1)
- Searching and encircling prey:
- (2)
- Spirally updating location:
2.3.10. Tune the Parameters of the Hybrid Machine Learning Models
2.4. Statistical Indicators
- Determination Coefficient (R2)
- Root Mean Square Error (RMSE)
- Mean Absolute Error (MAE)
- Normalized Root Mean Square Error (NRMSE)
- Percent of Bias (PBIAS)
3. Results
3.1. Linear Regression (LR) Model
3.2. ML Models with Single VI as Input
3.3. ML Models with All Features as Input
3.4. ML Models with Optimized Features as Input
3.5. Mapping AGB at Field Scale
4. Discussion
4.1. Uncertainty of Observed Data
4.2. Comparison of Different Models
4.3. Determining the Most Important VIs
4.4. Effect of Growth Stage on VIs-AGB Model
5. Conclusions
Author Contributions
Funding
Data Availability Statement
Conflicts of Interest
References
- Han, L.; Yang, G.; Dai, H.; Xu, B.; Yang, H.; Feng, H.; Li, Z.; Yang, X. Modeling maize above-ground biomass based on machine learning approaches using UAV remote-sensing data. Plant Methods 2019, 15, 10. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Kumar, L.; Mutanga, O. Remote sensing of above-ground biomass. Remote Sens. 2017, 9, 935. [Google Scholar] [CrossRef] [Green Version]
- Zhou, Y.; Xiao, X.; Qin, Y.; Dong, J.; Zhang, G.; Kou, W.; Li, X. Mapping paddy rice planting area in rice-wetland coexistent areas through analysis of Landsat 8 OLI and MODIS images. Int. J. Appl. Earth Obs. Geoinf. 2016, 46, 1–12. [Google Scholar] [CrossRef] [Green Version]
- Jin, X.; Li, Z.; Feng, H.; Ren, Z.; Li, S. Deep neural network algorithm for estimating maize biomass based on simulated Sentinel 2A vegetation indices and leaf area index. Crop J. 2020, 8, 87–97. [Google Scholar] [CrossRef]
- Dong, T.; Liu, J.; Qian, B.; He, L.; Liu, J.; Wang, R.; Shang, J. Estimating crop biomass using leaf area index derived from Landsat 8 and Sentinel-2 data. ISPRS J. Photogramm. Remote Sens. 2020, 168, 236–250. [Google Scholar] [CrossRef]
- Zhang, L.X.; Chen, Y.Q.; Li, Y.X.; Du, K.M.; Zheng, F.X.; Sun, Z.F. Estimating Above Ground Biomass of Winter Wheat at Early Growth Stages Based on Visual Spectral. Spectrosc. Spectr. Anal. 2019, 39, 2501. [Google Scholar]
- Yue, J.; Yang, G.; Tian, Q.; Feng, H.; Xu, K.; Zhou, C. Estimate of winter-wheat above-ground biomass based on UAV ultrahigh-ground-resolution image textures and vegetation indices. ISPRS J. Photogramm. Remote Sens. 2019, 150, 226–244. [Google Scholar] [CrossRef]
- Jia, M.; Li, W.; Wang, K.; Zhou, C.; Cheng, T.; Tian, Y.; Zhu, Y.; Cao, W.; Yao, X. A newly developed method to extract the optimal hyperspectral feature for monitoring leaf biomass in wheat. Comput. Electron. Agric. 2019, 165, 104942. [Google Scholar] [CrossRef]
- Geng, L.; Che, T.; Ma, M.; Tan, J.; Wang, H. Corn Biomass Estimation by Integrating Remote Sensing and Long-Term Observation Data Based on Machine Learning Techniques. Remote Sens. 2021, 13, 2352. [Google Scholar] [CrossRef]
- Liu, S.; Zeng, W.; Wu, L.; Lei, G.; Chen, H.; Gaiser, T.; Srivastava, A.K. Simulating the Leaf Area Index of Rice from Multispectral Images. Remote Sens. 2021, 13, 3663. [Google Scholar] [CrossRef]
- Wang, M.; Wan, Y.; Ye, Z.; Lai, X. Remote sensing image classification based on the optimal support vector machine and modified binary coded ant colony optimization algorithm. Inf. Sci. 2017, 402, 50–68. [Google Scholar] [CrossRef]
- Dong, J.; Liu, X.; Huang, G.; Fan, J.; Wu, L.; Wu, J. Comparison of four bio-inspired algorithms to optimize KNEA for predicting monthly reference evapotranspiration in different climate zones of China. Comput. Electron. Agric. 2021, 186, 106211. [Google Scholar] [CrossRef]
- Li, Z.; Zhao, Y.; Taylor, J.; Gaulton, R.; Jin, X.; Song, X.; Yang, G. Comparison and transferability of thermal, temporal and phenological-based in-season predictions of above-ground biomass in wheat crops from proximal crop reflectance data. Remote Sens. Environ. 2022, 273, 112967. [Google Scholar] [CrossRef]
- Zadoks, J.C.; Chang, T.T.; Konzak, C.F. A decimal code for the growth stages of cereals. Weed Res. 1974, 14, 415–421. [Google Scholar] [CrossRef]
- Buschmann, C.; Nagel, E. In vivo spectroscopy and internal optics of leaves as basis for remote sensing of vegetation. Int. J. Remote Sens. 1993, 14, 711. [Google Scholar] [CrossRef]
- Tucker, C.J. Red and photographic infrared linear combinations for monitoring vegetation. Remote Sens. Environ. 1979, 8, 127. [Google Scholar] [CrossRef] [Green Version]
- Gitelson, A.A.; Kaufman, Y.J.; Merzlyak, M.N. Use of a green channel in remote sensing of global vegetation from EOS-MODIS. Remote Sens. Environ. 1996, 58, 289–298. [Google Scholar] [CrossRef]
- Cao, Q.; Miao, Y.; Wang, H.; Huang, S.; Cheng, S.; Khosla, R.; Jiang, R. Non-destructive estimation of rice plant nitrogen status with Crop Circle multispectral active canopy sensor. Field Crops Res. 2013, 154, 133. [Google Scholar] [CrossRef]
- Gitelson, A.A. Remote estimation of canopy chlorophyll content in crops. Geophys. Res. Lett. 2005, 32, L08403. [Google Scholar] [CrossRef] [Green Version]
- Sripada, R.P.; Heiniger, R.W.; White, J.G.; Meijer, A.D. Aerial color infrared photography for determining early in-season nitrogen requirements in corn. Agron. J. 2006, 98, 968. [Google Scholar] [CrossRef]
- Lu, J.; Miao, Y.; Shi, W.; Li, J.; Yuan, F. Evaluating different approaches to non-destructive nitrogen status diagnosis of rice using portable RapidSCAN active canopy sensor. Sci. Rep. 2017, 7, 14073. [Google Scholar] [CrossRef] [PubMed]
- Haboudane, D.; Miller, J.R.; Pattey, E.; Zarco-Tejada, P.J.; Strachan, I.B. Hyperspectral vegetation indices and novel algorithms for predicting green LAI of crop canopies: Modeling and validation in the context of precision agriculture. Remote Sens. Environ. 2004, 90, 337–352. [Google Scholar] [CrossRef]
- Rouse, J.W., Jr.; Haas, R.H.; Schell, J.A.; Deering, D.W. Monitoring vegetation systems in the Great Plains with ERTS. In NASA. Goddard Space Flight Center 3d ERTS-1 Symphony; NASA: Washington, DC, USA, 1974; p. 309. [Google Scholar]
- Jordan, C.F. Derivation of leaf-area index from quality of light on the forest floor. Ecology 1969, 50, 663–666. [Google Scholar] [CrossRef]
- Roujean, J.; Breon, F. Estimating PAR absorbed by vegetation from bidirectional reflectance measurements. Remote Sens. Environ. 1995, 51, 375–384. [Google Scholar] [CrossRef]
- Huete, A.R. A soil-adjusted vegetation index (SAVI). Remote Sens. Environ. 1988, 25, 295. [Google Scholar] [CrossRef]
- Sandham, L. Surface temperature measurement from space: A case study in the south western cape of south Africa. S. Afr. J. Enol. Vitic. 1997, 18, 25. [Google Scholar] [CrossRef] [Green Version]
- Chen, J.M. Evaluation of vegetation indices and a modified simple ratio for boreal applications. Can. J. Remote Sens. 1996, 22, 229. [Google Scholar] [CrossRef]
- Reyniers, M.; Walvoort, D.J.; De Baardemaaker, J. A linear model to predict with a multi-spectral radiometer the amount of nitrogen in winter wheat. Int. J. Remote Sens. 2006, 27, 4159–4179. [Google Scholar] [CrossRef]
- Dash, J.; Curran, P. The MERIS terrestrial chlorophyll index. Int. J. Remote Sens. 2004, 25, 5403–5413. [Google Scholar] [CrossRef]
- Goel, N.S.; Qin, W. Influences of canopy architecture on relationships between various vegetation indices and LAI and FPAR: A computer simulation. Remote Sens. Environ. 1994, 10, 309–347. [Google Scholar] [CrossRef]
- Gong, P.; Pu, R.; Biging, G.S.; Larrieu, M.R. Estimation of forest leaf area index using vegetation indices derived from Hyperion hyperspectral data. IEEE Geosci. Remote Sens. Lett. 2003, 41, 1355–1362. [Google Scholar] [CrossRef] [Green Version]
- Barnes, E.; Clarke, T.; Richards, S.; Colaizzi, P.; Haberland, J.; Kostrzewski, M.; Waller, P.; Choi, C.; Riley, E.; Thompson, T.; et al. Coincident detection of crop water stress, nitrogen status and canopy density using ground based multispectral data. In Proceedings of the Fifth International Conference on Precision Agriculture, Bloomington, MN, USA, 16–19 July 2000. [Google Scholar]
- Jasper, J.; Reusch, S.; Link, A. Active sensing of theNstatus of wheat using optimized wavelength combination: Impact of seed rate, variety and growth stage. Precis. Agric. 2009, 9, 23–30. [Google Scholar]
- Gitelson, A.A.; Gritz, Y.; Merzlyak, M.N. Relationships between leaf chlorophyll content and spectral reflectance and algorithms for non-destructive chlorophyll assessment in higher plant leaves. J. Plant Physiol. 2003, 160, 271–282. [Google Scholar] [CrossRef]
- Elsayed, S.; Rischbeck, P.; Schmidhalter, U. Comparing the performance of active and passive reflectance sensors to assess the normalized relative canopy temperature and grain yield of drought-stressed barley cultivars. Field Crops Res. 2015, 177, 148–160. [Google Scholar] [CrossRef]
- Erdle, K.; Mistele, B.; Schmidhalter, U. Comparison of active and passive spectral sensors in discriminating biomass parameters and nitrogen status in wheat cultivars. Field Crops Res. 2011, 124, 74. [Google Scholar] [CrossRef]
- Datt, B. Visible/near infrared reflectance and chlorophyll content in Eucalyptus leaves. Int. J. Remote Sens. 1999, 20, 2741–2759. [Google Scholar] [CrossRef]
- Dai, H.; MacBeth, C. Effects of learning parameters on learning procedure and performance of a BPNN. Neural Netw. 1997, 10, 1505–1521. [Google Scholar] [CrossRef] [Green Version]
- Quinonero-Candela, J.; Rasmussen, C.E. A unifying view of sparse approximate Gaussian process regression. J. Mach. Learn. Res. 2005, 6, 1939–1959. [Google Scholar]
- Vapnik, V.; Izmailov, R. Knowledge transfer in SVM and neural networks. Ann. Math. Artif. Intell. 2017, 81, 3–19. [Google Scholar] [CrossRef]
- Breiman, L. Random forests. Mach. Learn. 2001, 45, 5–32. [Google Scholar] [CrossRef] [Green Version]
- Friedman, J.H. Stochastic gradient boosting. Comput. Stat. Data Anal. 2002, 38, 367–378. [Google Scholar] [CrossRef]
- Chen, T.; Guestrin, C. Xgboost: A scalable tree boosting system. In Proceedings of the 22nd ACM Sigkdd International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA, 13–17 August 2016. [Google Scholar]
- Mafarja, M.; Aljarah, I.; Heidari, A.A.; Hammouri, A.I.; Faris, H.; Ala’M, A.Z.; Mirjalili, S. Evolutionary population dynamics and grasshopper optimization approaches for feature selection problems. Knowl.-Based Syst. 2018, 145, 25–45. [Google Scholar] [CrossRef] [Green Version]
- Yue, J.; Feng, H.; Jin, X.; Yuan, H.; Li, Z.; Zhou, C.; Yang, G.; Tian, Q. A comparison of crop parameters estimation using images from UAV-mounted snapshot hyperspectral sensor and high-definition digital camera. Remote Sens. 2018, 10, 1138. [Google Scholar] [CrossRef] [Green Version]
- Li, W.; Niu, Z.; Chen, H.; Li, D.; Wu, M.; Zhao, W. Remote estimation of canopy height and aboveground biomass of maize using high-resolution stereo images from a low-cost unmanned aerial vehicle system. Ecol. Indic. 2016, 67, 637–648. [Google Scholar] [CrossRef]
- Mirjalili, S.; Lewis, A. The whale optimization algorithm. Adv. Eng. Softw. 2016, 95, 51–67. [Google Scholar] [CrossRef]
- Jimenezberni, J.A.; Deery, D.M.; Rozaslarraondo, P.; Condon, A.G.; Rebetzke, G.J.; James, R.A.; Bovill, W.D.; Furbank, R.T.; Sirault, X.R.R. High throughput determination of plant height, ground cover, and above-ground biomass in wheat with LiDAR. Front. Plant Sci. 2018, 9, 237. [Google Scholar] [CrossRef] [Green Version]
- Chakraborty, A.; Goswami, D. Prediction of slope stability using multiple linear regression (MLR) and artificial neural network (ANN). Arab. J. Geosci. 2017, 10, 385. [Google Scholar] [CrossRef]
- Xu, W.; Chen, P.; Zhan, Y.; Chen, S.; Zhang, L.; Lan, Y. Cotton yield estimation model based on machine learning using time series UAV remote sensing data. Int. J. Appl. Earth Obs. Geoinf. 2021, 104, 102511. [Google Scholar] [CrossRef]
- Jannoura, R.; Joergensen, R.G.; Bruns, C. Organic fertilizer effects on growth, crop yield, and soil microbial biomass indices in sole and intercropped peas and oats under organic farming conditions. Eur. J. Agron. 2014, 52, 259–270. [Google Scholar] [CrossRef]
- Hunt, E.R.; Cavigelli, M.; Daughtry, C.S.; Mcmurtrey, J.E.; Walthall, C.L. Evaluation of digital photography from model aircraft for remote sensing of crop biomass and nitrogen status. Precis. Agric. 2005, 6, 359–378. [Google Scholar] [CrossRef]
- Bosch Serra, A.D.; Casanova, D. Estimation of onion (Allium cepa, L.) biomass and light interception from reflectance measurements at field level. In Proceedings of the XXV International Horticultural Congress, Part 9: Computers and Automation, Electronic Information in Horticulture, Brussels, Belgium, 2–7 August 1998; pp. 53–64. [Google Scholar]
- Alvino, A.; Marino, S. Remote sensing for irrigation of horticultural crops. Horticulturae 2017, 3, 40. [Google Scholar] [CrossRef] [Green Version]
- Marshall, M.; Belgiu, M.; Boschetti, M.; Pepe, M.; Stein, A.; Nelson, A. Field-level crop yield estimation with PRISMA and Sentinel-2. ISPRS J. Photogramm. Remote Sens. 2022, 187, 191–210. [Google Scholar] [CrossRef]
UAV | Description | Sensor | Description |
---|---|---|---|
Name | DJI-Phantom 4 pro | Type | FC6360 |
Flight altitude above ground level | 50 m | Bands | Blue (450 nm ± 16 nm), Green (560 nm ± 16 nm), Red (650 nm ± 16 nm), Red edge (730 nm ± 16 nm), NIR (840 nm ± 26 nm) |
Flight speed | 4 m/s | Number of images | 11,200 |
Satellite systems | GPS | Shutter speed | 1.2 |
Forward overlap | 80% | ISO sensibility | ISO-200 |
Side overlap | 80% | Image dimension | 1600 × 1300 |
Field of view | 90° | Resolution | 5.3 cm/pixel |
Shooting interval | 2 s | Image format | JGEG, TIFF |
No. | VI | Formula | Reference |
---|---|---|---|
1 | Green Ratio Vegetation Index (GRVI) | GRVI = NIR1/G | Buschmann and Nagel (1993) [15] |
2 | Green Difference Vegetation Index (GDVI) | GDVI = NIR − G | Tucker (1979) [16] |
3 | Green Normalized Difference Vegetation Index (GNDVI) | GNDVI = (NIR − G)/(NIR + G) | Gitelson and Kaufman (1996) [17] |
4 | Green Wide Dynamic Range Vegetation Index (GWDRVI) | GWDRVI = (0.12 × NIR − G)/(0.12 × NIR + G) | Cao et al. (2013) [18] |
5 | Green Chlorophyll Index (CIg) | Cig = NIR/G − 1 | Gitelson (2005) [19] |
6 | Modified Green Simple Ratio (MSR_G) | MSR_G = (NIR/G − 1)/sqrt(NIR/G + 1) | Cao et al. (2013) [18] |
7 | Green Soil-Adjusted Vegetation Index (GSAVI) | GSAVI = 1.5 × (NIR − G)/(NIR + G + 0.5) | Sripada et al. (2016) [20] |
8 | Green Re-normalized Different Vegetation Index (GRDVI) | GRDVI = (NIR − G)/sqrt(NIR + G) | Cao et al. (2013) [18] |
9 | Normalized Green Index (NGI) | NGI = G/(NIR + G + RE) | Sripada et al. (2016) [20] |
10 | Normalized Red Edge Index (NREI) | NREI = RE/(NIR + G + RE) | Cao et al. (2013) [18] |
11 | Normalized Red Index (NRI) | NRI = R/(NIR + R + RE) | Lu et al. (2014) [21] |
12 | Normalized NIR Index (NNIR) | NNIR = NIR/(NIR + R + RE) | Sripada et al. (2016) [20] |
13 | Modified Double Difference Index (MDD) | MDD = (NIR − RE)/(RE − G) | Lu et al. (2014) [21] |
14 | Modified Normalized Difference Index (MNDI) | MNDI = (NIR − RE)/(NIR − G) | Cao et al. (2013) [18] |
15 | Modified Enhanced Vegetation Index (MEVI) | MEVI = 2.5 × (NIR − RE)/(NIR + 6 × RE − 7.5 × G + 1) | Cao et al. (2013) [18] |
16 | Modified Normalized Difference Red Edge (MNDRE) | MNDRE = (NIR − RE − 2 × G)/(NIR + RE − 2 × G) | Cao et al. (2013) [18] |
17 | Modified Chlorophyll Absorption In Reflectance Index 1 (MCARI1) | MCARI1 = ((NIR − RE) − 0.2 × (NIR − R)) × (NIR/RE) | Haboudane et al. (2004) [22] |
18 | Modified Chlorophyll Absorption In Reflectance Index 2 (MCARI2) | MCARI2 = 1.5 × (2.5 × (NIR − R) − 1.3 × (NIR − RE))/sqrt((2 × NIR + 1)2 − (6 × NIR − 5 × sqrt(R) − 0.5) | Haboudane et al. (2004) [22] |
19 | Normalized Difference Vegetation Index (NDVI) | NDVI = (NIR − R)/(NIR + R) | Rouse et al. (1974) [23] |
20 | Ratio Vegetation Index(RVI) | RVI = NIR/R | Jordan et al. (1969) [24] |
21 | Difference Vegetation Index (DVI) | DVI = NIR − R | Tucker (1979) [16] |
22 | Renormalized Difference Vegetation Index (RDVI) | RDVI = (NIR − R)/sqrt(NIR + R) | Roujean and Breon (1995) [25] |
23 | Wide Dynamic Range Vegetation Index (WDRVI) | WDRVI = (0.12 × NIR − R)/(0.12 × NIR + R) | Gitelson et al. (2004) [19] |
24 | Soil-Adjusted Vegetation Index (SAVI) | SAVI = 1.5 × (NIR − R)/(NIR + R + 0.5) | Huete et al. (1988) [26] |
25 | Transformed Normalized Vegetation Index (TNDVI) | TNDVI = sqrt((NIR − R)/(NIR + R) + 0.5) | Sandham (1997) [27] |
26 | Modified Simple Ratio (MSR) | MSR = (NIR/R − 1)/sqrt(NIR/R + 1) | Chen (1996) [28] |
27 | Optimal Vegetation Index (VIopt) | VIopt = 1.45 × (NIR^2 + 1)/(R + 0.45) | Reyniers et al. (2006) [29] |
28 | MERIS Terrestrial Chlorophyll Index (MTCI) | MTCI = (NIR − RE)/(RE − R) | Dash and Curran (2004) [30] |
29 | Nonlinear Index (NLI) | NLI = (NIR^2 − R)/(NIR^2 + R) | Goel and Qin (1994) [31] |
30 | Modified Nonlinear Index (MNLI) | MNLI = 1.5 × (NIR^2 − R)/(NIR^2 + R + 0.5) | Gong et al. (2003) [32] |
31 | NDVI × RVI | NDVI_RVI = (NIR^2 − R)/(NIR + R^2) | Gong et al. (2003) [32] |
32 | SAVI × SR | SAVI_SR = (NIR^2 − R)/(NIR + R + 0.5) × R | Gong et al. (2003) [32] |
33 | Normalized Difference Red Edge (NDRE) | NDRE = (NIR − RE)/(NIR + RE) | Barnes et al. (2000) [33] |
34 | Red Edge Ratio Vegetation Index (RERVI) | RERVI = (NIR/RE) | Gitelson et al. (1996) [17] |
35 | Red Edge Difference Vegetation Index (REDVI) | REDVI = (NIR − RE) | Cao et al. (2013) [18] |
36 | Red Edge Re-normalized Different Vegetation Index (RERDVI) Red Edge Wide Dynamic Range Vegetation Index (REWDRVI) Red Edge Soil-Adjusted Vegetation Index (RESAVI) | REWDRVI = (0.12 × NIR − R)/(0.12 × NIR + R) | Cao et al. (2013) [18] |
37 | Red Edge Optimal Soil-Adjusted Vegetation Index (REOSAVI) | REOSAVI = 1.5 × (NIR − RE)/(NIR + RE + 0.5) | Cao et al. (2013) [18] |
38 | Optimized Red Edge Vegetation Index (REVIopt) | REVIopt = 100 (log(NIR) − log(RE)) | Jasper et al. (2009) [34] |
39 | Red Edge Chlorophyll Index (CIre) | CIre = NIR/RE − 1 | Gitelson et al. (2003) [35] |
40 | Modified Red Edge Simple Ratio (MSR_RE) | MSR_RE = (NIR/RE − 1)/sqrt(NIR/RE + 1) | Lu et al. (2014) [21] |
41 | Red Edge Normalized Difference Vegetation Index (RENDVI) | RENDVI = (NIR − RE)/(NIR + RE) | Elsayed et al. (2015) [36] |
42 | Red Edge Simple Ratio (RESR) | RESR = RE/R | Erdle et al. (2011) [37] |
43 | Modified Red Edge Difference Vegetation Index (MREDVI) MERIS Terrestrial Chlorophyll Index (MTCI) | MREDVI = RE −R | Cao et al. (2013) [18] |
44 | DATT Index (DATT) | DATT = (NIR − RE)/(NIR − R) | Datt (1999) [38] |
45 | Normalized Near-Infrared Index (NNIRI) | NNIRI = NIR/(NIR + RE + R) | Lu et al. (2014) [21] |
46 | Normalized Red Edge Index (NREI) | NREI = RE/(NIR + RE + R) | Lu et al. (2014) [21] |
47 | Normalized Red Index (NRI) | NRI = R/(NIR + RE + R) | Lu et al. (2014) [21] |
48 | Modified Double Difference Index (MDD) | MDD_R = NIR − R | Lu et al. (2014) [21] |
49 | Modified Red Edge Simple Ratio (MRESR) | MRESR = (NIR − R)/(RE − R) | Lu et al. (2014) [21] |
50 | Modified Normalized Difference Index (MNDI) | MNDI = (NIR − RE)/(NIR + RE − 2 × R) | Lu et al. (2014) [21] |
51 | Modified Enhanced Vegetation Index (MEVI) | MEVI_R = 2.5 × (NIR − RE)/(NIR + 6 × RE − 7.5 × R + 1) | Lu et al. (2014) [21] |
52 | Modified Normalized Difference Red Edge (MNDRE2) | MNDRE2 = (NIR − RE + 2 × R)/(NIR + RE − 2 × R) | Lu et al. (2014) [21] |
53 | Red Edge Transformed Vegetation Index (RETVI) | RETVI = 0.5 × (120 × (NIR − R) − 200 × (RE − R)) | Lu et al. (2014) [21] |
54 | Modified Chlorophyll Absorption In Reflectance Index 3 (MCARI3) | MCARI3 = ((NIR − RE) − 0.2 × (NIR − R)) × (NIR/RE) | Haboudane et al. (2004) [22] |
55 | Modified Chlorophyll Absorption In Reflectance Index 4 (MCARI4) | MCARI4 = (1.5 × (2.5 × (NIR − G) − 1.3 × (NIR − RE))/(sqrt((2 × NIR + 1)^2 − (6 × NIR − 5 × sqrt(G)) − 0.5)) | Haboudane et al. (2004) [22] |
56 | Modified Red Edge Transformed Vegetation Index (MRETVI) Modified Canopy Chlorophyll Content Index (MCCCI) | MRETVI = 1.2 × (1.2 × (NIR − R) − 2.5 × (RE − R)) | Lu et al. (2014) [21] |
Wheat AGB | Max | Min | Mean | Median | Std. | CV | Skewness |
---|---|---|---|---|---|---|---|
Training | 2.31 | 0.25 | 1.16 | 1.15 | 1.10 | 0.49 | 0.21 |
Testing | 2.11 | 0.14 | 1.10 | 1.04 | 1.00 | 0.54 | 0.28 |
ID | Regression Equation | RMSE | R2 | MAE | NRMSE | PBIAS |
---|---|---|---|---|---|---|
MLR | ||||||
LR | 0.286 + 0.152 × G | 0.851 | 0.383 | 0.693 | 0.399 | 0 |
MLR1 | 0.760 + 0.088 × G − 2.04 × CIg | 0.842 | 0.396 | 0.687 | 0.395 | 0 |
MLR2 | 14.957 + 0.3550G + 31.28 × Cig − 2.312 × MSR_G | 0.796 | 0.461 | 0.626 | 0.373 | 0 |
MLR3 | −5.306 − 0.304 × G − 21.125 × Cig + 0.151 × MSR_G + 0.022 × GNDVI | 0.796 | 0.461 | 0.626 | 0.373 | 0 |
GOA-MLR | ||||||
GOA-LR | 0.286 + 0.152 × G | 0.851 | 0.383 | 0.693 | 0.399 | 0 |
GOA-MLR1 | −1.326 − 1.1606 × MSR_G 1.506 × NGI | 0.718 | 0.560 | 0.564 | 0.337 | 0 |
GOA-MLR2 | 0.891 − 1.205CIg + 1.028 × GRDVI + 0.027 × MCARI3 | 0.67 | 0.587 | 0.553 | 0.327 | 0 |
GOA-MLR3 | 2.591 − 1.017GDVI + 0.023 × CIre + 0.07835047 × GNDVI + 0.001 × NGI | 0.701 | 0.582 | 0.558 | 0.328 | 0 |
Model/VI | RMSE (kg m−2) | R2 | MAE (kg m−2) | NRMSE | PBIAS (kg m−2) |
---|---|---|---|---|---|
MLP | |||||
NIR | 0.375 | 0.631 | 0.293 | 0.205 | 0.052 |
GRVI | 0.465 | 0.443 | 0.364 | 0.253 | 0.072 |
NREI | 0.473 | 0.411 | 0.386 | 0.258 | 0.062 |
GPR | |||||
NIR | 0.318 | 0.725 | 0.267 | 0.173 | 0.020 |
MSR_G | 0.348 | 0.701 | 0.283 | 0.190 | 0.068 |
GDVI | 0.348 | 0.701 | 0.283 | 0.190 | 0.068 |
SVM | |||||
NIR | 0.349 | 0.677 | 0.278 | 0.190 | 0.050 |
CIg | 0.377 | 0.654 | 0.312 | 0.205 | 0.074 |
MSR_G | 0.377 | 0.669 | 0.310 | 0.205 | 0.085 |
RF | |||||
NIR | 0.391 | 0.642 | 0.307 | 0.213 | 0.029 |
CIg | 0.413 | 0.583 | 0.322 | 0.225 | 0.043 |
GWDRVI | 0.413 | 0.581 | 0.322 | 0.225 | 0.043 |
GBDT | |||||
NIR | 0.382 | 0.643 | 0.300 | 0.208 | 0.029 |
CIg | 0.401 | 0.599 | 0.320 | 0.219 | 0.055 |
GWDRVI | 0.402 | 0.595 | 0.321 | 0.219 | 0.053 |
XGB | |||||
NIR | 0.327 | 0.708 | 0.267 | 0.178 | −0.036 |
CIg | 0.339 | 0.679 | 0.281 | 0.185 | −0.016 |
MSR_G | 0.339 | 0.679 | 0.281 | 0.185 | −0.016 |
Model | RMSE (kg m−2) | R2 | MAE (kg m−2) | NRMSE | PBIAS (kg m−2) |
---|---|---|---|---|---|
MLP1 | 0.334 | 0.722 | 0.256 | 0.182 | 0.064 |
GPR1 | 0.276 | 0.801 | 0.214 | 0.151 | 0.009 |
SVM1 | 0.301 | 0.747 | 0.223 | 0.164 | −0.015 |
PSO-SVM1 | 0.299 | 0.750 | 0.220 | 0.162 | 0.011 |
WOA-SVM1 | 0.291 | 0.751 | 0.218 | 0.162 | 0.201 |
GOA-SVM1 | 0.298 | 0.752 | 0.239 | 0.162 | 0.015 |
RF1 | 0.264 | 0.815 | 0.195 | 0.144 | 0.035 |
GBDT1 | 0.271 | 0.808 | 0.196 | 0.148 | 0.038 |
XGBoost1 | 0.246 | 0.854 | 0.187 | 0.134 | −0.042 |
PSO-XGB1 | 0.247 | 0.84 | 0.192 | 0.224 | 0.175 |
WOA-XGB1 | 0.240 | 0.842 | 0.182 | 0.218 | 0.165 |
GOA-XGB1 | 0.232 | 0.847 | 0.178 | 0.127 | −0.004 |
Model | RMSE (kg m−2) | R2 | MAE (kg m−2) | NRMSE | PBIAS (kg m−2) |
---|---|---|---|---|---|
MLP2 | 0.355 | 0.647 | 0.281 | 0.193 | 0.0258 |
GPR2 | 0.256 | 0.826 | 0.201 | 0.140 | 0.015 |
SVM2 | 0.288 | 0.770 | 0.219 | 0.157 | −0.001 |
PSO-SVM2 | 0.287 | 0.780 | 0.226 | 0.261 | 0.206 |
WOA-SVM2 | 0.287 | 0.771 | 0.222 | 0.263 | 0.201 |
GOA-SVM2 | 0.284 | 0.771 | 0.217 | 0.155 | −0.001 |
RF2 | 0.253 | 0.831 | 0.190 | 0.138 | 0.032 |
GBDT2 | 0.272 | 0.805 | 0.194 | 0.148 | 0.035 |
XGBoost2 | 0.243 | 0.858 | 0.181 | 0.133 | −0.043 |
PSO-XGB2 | 0.249 | 0.842 | 0.193 | 0.226 | 0.175 |
WOA-XGB2 | 0.236 | 0.849 | 0.179 | 0.214 | 0.162 |
GOA-XGB2 | 0.226 | 0.855 | 0.172 | 0.123 | −0.001 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Han, Y.; Tang, R.; Liao, Z.; Zhai, B.; Fan, J. A Novel Hybrid GOA-XGB Model for Estimating Wheat Aboveground Biomass Using UAV-Based Multispectral Vegetation Indices. Remote Sens. 2022, 14, 3506. https://doi.org/10.3390/rs14143506
Han Y, Tang R, Liao Z, Zhai B, Fan J. A Novel Hybrid GOA-XGB Model for Estimating Wheat Aboveground Biomass Using UAV-Based Multispectral Vegetation Indices. Remote Sensing. 2022; 14(14):3506. https://doi.org/10.3390/rs14143506
Chicago/Turabian StyleHan, Yixiu, Rui Tang, Zhenqi Liao, Bingnian Zhai, and Junliang Fan. 2022. "A Novel Hybrid GOA-XGB Model for Estimating Wheat Aboveground Biomass Using UAV-Based Multispectral Vegetation Indices" Remote Sensing 14, no. 14: 3506. https://doi.org/10.3390/rs14143506