Table 5 Image features extracted from lung nodule images
From: Semantic characteristic grading of pulmonary nodules based on deep neural networks
Haralick features calculated from gray-level co-occurrence matrices, such as Autocorrelation, Contrast, Correlation, Cluster Prominence, Cluster Shade, Dissimilarity, Energy, Entropy, Homogeneity, Maximum probability, Sum of squares, Sum average, Sum variance, Sum entropy, Difference variance, Difference entropy and so on | |
Image entropy | |
Geometric features | Centroid, MajorAxisLength, MinorAxisLength, Eccentricity, Orientation, ConvexArea, FilledArea, EulerNumber, EquivDiameter, EquivDiameter, Solidity, Extent, Perimeter, PerimeterOld |
Hu's invariant moments [33] | |
Intensity features | mean, variance, skewness |