Steps: Feature Extraction from Input Images Using the VGG16 Model.
Input:	BreakHis 400× training image data
Preprocessing:	Resizing the images to 512 × 512
	Normalizing pixel values to between 0 and 1
Output:	Initialize the VGG16 model.
	Import VGG16 model
	Remove the fully connected layers.
	Load the VGG16 model for feature extraction by removing the FC layers as they were designed for ImageNet classification tasks.
	VGG_model = VGG16 (weights = ’Imagenet’, include_top = False,
	input_shape = (512, 512, 3)
	3-is the RGB channel since we are using color images
	Pass the input images through the model:
	Forward pass the preprocessed images through the VGG16 model to obtain the output feature maps.
	Retrieve the output of the last convolutional layer as they capture the high-level abstract features.
	Flatten the features.
	Flatten(img): Convert the 3-D extracted features to the 1-D feature vector
	Store the extracted features with corresponding image labels for classification tasks
	Features(img) = Flatten(img)