Skip to main content

Table 3 Summary of selected references applying hyperspectral imaging to seed quality defect detection

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

Seed

Spectral rangea

Varieties

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

Mung bean

900–1700 (1000–1600)

1 variety, 8 treatments

2400

Spectra and image

PCA

Reflectance

OWb (single kernels)

LDA, QDA

Insect damage detection

> 82%

Kaliramesh et al. [38]

Soybean

900–1700 with soft x-ray

1 variety, 5 treatments

1000

Spectra and image

GLCM

Reflectance

OW (single kernels)

LDA, QDA

Insect damage detection

99% (QDA)

Chelladurai et al. [39]

Wheat

700–1100

1 variety, 4 insect varieties

1500

Spectra and image

STEPDISC, GLCM, GLRM, PCA

Reflectance

OW (single kernels)

LDA, QDA

Insect damage detection

95.3–99.3%

Singh et al. [37]

Wheat

400–1000 (450–920)

1 variety, 3 treatments

144

Spectra and image

PCA

Reflectance

PWc prediction map and OW (single kernels)

Spectral index

Seed sprouted detection

> 90%

Xing et al. [36]

  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