Table 3 Experimental setup
From: Machine learning hyperparameter selection for Contrast Limited Adaptive Histogram Equalization
Element | Choice |
|---|---|
Datasets | Dataset 1: 246 contrast distorted images |
Dataset 2: 6258 contrast distorted images | |
Full-reference IQA techniques | Mean squared error (MSE) |
Peak signal-to-noise ratio (PSNR) | |
Structural similarity (SSIM) | |
Gray-level entropy difference (GLED) | |
Absolute mean brightness error (AMBE) | |
Visual information fidelity (VIF) | |
Image features | Spatial (3 features) |
Histogram (3 features) | |
Texture (19 features) | |
Image quality (3 features) | |
Supervised regression algorithms | Classification and regression tree (CART) |
Multi-layer perceptron (MLP) | |
Support vector machine (SVM) | |
Random forest (RF) | |
Extreme Gradient Boosting (XGBoost) | |
Evaluation measures | Root mean square error (RMSE) |
R-squared (R2) | |
Resampling method | 10-fold cross-validation |