Skip to main content

Table 6 Summary of selected references applying hyperspectral imaging to seed composition

From: Hyperspectral imaging for seed quality and safety inspection: a review

Seed Spectral rangea Components Sample numbers Features Signal mode Data analysis strategies Main application type Classification result (highest accuracy) References
Spectra/image Extraction/selection methods Analysis level Classification/regression methods
Grape seed 865–1712 (977–1625) Total iron-reactive phenolics, anthocyanins and tannins 60 Spectra No Reflectance OWb (single kernels) PCR, PLSR, SVR Regression R2 = 0.91, 0.88, 0.90, respectively Zhang et al. [67]
Grape seed 884–1717 (950–1650) Flavanol content 99 Spectra PCA, PLSR loadings Reflectance OW (bulk), PWc PLSR Regression R2 = 0.88 Rodríguez-Pulido et al. [72]
Maize 700–1100 (750–1090) Moisture and oil 473 for moisture, 151 for oil Spectra GA-PLS Absorbance OW (single kernels) PLS Regression SECV = 1.20%, 1.38%, respectively Cogdill et al. [75]
Maize 310–1100 (400–1000) Hardness, springiness, and resilience 252 Spectra SPA Reflectance OW (single), PW prediction map PLSR Regression R2 = 0.90, 0.87, 0.85, respectively Wang et al. [69]
Maize 950–1700 Oil and oleic acid 400 Spectra GA-PLS Reflectance OW (single kernels), PW prediction map PLS Regression RMSEP = 0.7%, 14%, respectively Weinstock et al. [76]
Maize 960–1662 (1000–2498) Hard, intermediate and soft 36 Spectra PCA Reflectance OW (single kernels), PW PCA and prediction map, PLS-wise spectra model PLS-DA Classification 98% William et al. [70]
Rice 871–1766 Moisture 120 Spectra PCA, SPA Reflectance OW (bulk) SVR, LS-SVR and bacterial colony chemotaxis LS-SVR (BCC-LS-SVR) Regression R2 = 0.98 (BCC-LS-SVR) Sun et al. [71]
Wheat 980–2500 (1060–2500) Protein 4150 Spectra PLS Reflectance OW (single kernels), prediction map PLS Regression R2 = 0.82 Caporaso et al. [74]
Wheat 1000–2500 (1235–2450) Alpha-amylase activities 264 Spectra No Reflectance OW (single kernels) PLSR Regression R2 = 0.88 Xing et al. [68]
Wheat 860–1700 Hardness and protein 7200 Spectra PLSR, PCR Reflectance OW (bulk) PLSR, PCR Regression R2 = 0.88 Mahesh et al. [73]
Wheat 865–1711 (928–1695) Protein 79 Spectra No Reflectance OW (bulk) PCR, PLSR, RBFNN Regression R2 = 0.92 Yang et al. [7]
  1. aThe spectral range without brackets relates to the range acquisition of instrument, while the spectral range in brackets represents the spectral range for practical analysis
  2. bOW means objective-wise analysis, which means the analysis on ROIs (ROI can be bulk, single kernel or self-defined)
  3. cPW means pixel-wise analysis, which is the analysis on the pixels