Table 1. Pseudocode for proposed algorithms.

pseudocode for proposed algorithms
Data Loading
1:     shenzhen_data = load_images(shenzhen_path)
2:     montgomery_data = load_images(montgomery_path)
3:     combine_datasets(shenzhen_data, montgomery_data)
4:     train_data, val_data, test_data = load_data("combined_data ")
5:     return train_test_split(combined_data, test_size=0.2)
Training and Evaluation of Models
6:     for model_type in ["U-Net","Prop.U-Net","V-Net","Prop. V-Net","Seg-Net"]:
7:            Training and evaluation: {model_type}
8:            model = create_model(model_type)
9:            models = {build_unet(input_shape),build_prop_unet(input_shape),build_vnet(input_shape), build_prop_vnet(input_shape),                    "Seg-Net": build_segnet(input_shape)}
Model Results
10:    history, evaluation = train_and_evaluate_model(model, train_data, val_data)
11:    ResultofModel: {evaluation}") metrics=["IOU", "dice_coefficient"])
Prediction and Visualization on Test Data
12:    predict_and_visualize(model, test_data)