preprocessing_image.py

"""

Sorbonne University - Barkicay university

Master of enginerring sciences - Intelligent Systems Engineering (ISI) 

Project of Master 1. Preporcessing Rectal Tumor MR images


file: Preprocessing image

"""

# Import Librairies
import numpy as np
import cv2
import os
from matplotlib import pyplot as plt
import random


IMG_SIZE = 224
np.random.seed(42)

# Function for putting the data in folder and sperate them between data et label
def load_images_from_folder(folder, target_size=(224, 224)):
    images = []
    labels = []
    label_map = {'benign': 0, 'malignant': 1} # Mapping textual label by 1 or 0 label
    for label in os.listdir(folder):
        label_path = os.path.join(folder, label)
        if os.path.isdir(label_path):
            for filename in os.listdir(label_path):
                img_path = os.path.join(label_path, filename)
                img = cv2.imread(img_path, cv2.IMREAD_GRAYSCALE)  # download the image by grayscale
                if img is not None:
                    images.append(img)
                    labels.append(label_map.get(label, 1)) # label the image
    return images, labels


###########################################################################

# Function to preprocess image data by croping and fit for CNN
def crop_and_resize(image_path, target_size=(IMG_SIZE,IMG_SIZE)):
    
    image = cv2.imread(image_path)
    height, width = image.shape[:2]
    
    # Calculate the coordinates for the center crop
    center_x, center_y = width // 2, height // 2
    crop_size = min(height, width)
    x1 = center_x - crop_size // 2
    y1 = center_y - crop_size // 2
    x2 = x1 + crop_size
    y2 = y1 + crop_size
    
    # Crop the image
    cropped_image = image[y1:y2, x1:x2]
    
    # Resize the image
    resized_image = cv2.resize(cropped_image, target_size, interpolation=cv2.INTER_AREA)
    
    cv2.imwrite(image_path, resized_image)


def affiche(history):
    
    # summarize history for accuracy
    plt.subplot(211)
    plt.plot(history.history['accuracy'])
    plt.plot(history.history['val_accuracy'])
    plt.title('model accuracy')
    plt.ylabel('accuracy')
    plt.xlabel('epoch')
    plt.legend(['train', 'test'], loc='upper left')
    plt.grid()
    plt.show()


    # summarize history for loss
    plt.subplot(212)
    plt.plot(history.history['loss'])
    plt.plot(history.history['val_loss'])
    plt.title('model loss')
    plt.ylabel('loss')
    plt.xlabel('epoch')
    plt.legend(['train', 'test'], loc='upper left')
    plt.grid()
    plt.show()
    
##############################################################################

#####################################################################################

# Apply CLAHE to a grayscale image
def apply_clahe_grayscale(image_path):
    
    # Read the image in grayscale
    img = cv2.imread(image_path, cv2.IMREAD_GRAYSCALE)
    
    # Check if the image has been loaded correctly
    if img is None:
        print(f"Error loading image: {image_path}")
        return

    # Apply CLAHE on the grayscale image
    clahe = cv2.createCLAHE(clipLimit=2.0, tileGridSize=(8, 8))
    cl_img = clahe.apply(img)
    
    # Save the modified image (overwrite the original)
    cv2.imwrite(image_path, cl_img)

#####################################################################################


##############################################################################

# Function applying sobel filter
def apply_sobel_filter(image_path):
    # Read the image in grayscale
    image = cv2.imread(image_path, cv2.IMREAD_GRAYSCALE)
    
    # Apply sobel filter (kernel 5x5)
    sobelx = cv2.Sobel(image, cv2.CV_64F, 1, 0, ksize=5)
    sobely = cv2.Sobel(image, cv2.CV_64F, 0, 1, ksize=5)
    
    # Compute the magnitude
    sobel_image = cv2.magnitude(sobelx, sobely)
    
    # Normalize the image to range 0-255 and convert to uint8
    sobel_image = cv2.normalize(sobel_image, None, 0, 255, cv2.NORM_MINMAX)
    sobel_image = np.uint8(sobel_image)
    
    # Save the modified image (overwrite the original)
    cv2.imwrite(image_path, sobel_image)
    
# Function for Gaussian filter preprocessing
def apply_gaussian_filter(image_path):
    # Read the image in grayscale
    image = cv2.imread(image_path, cv2.IMREAD_GRAYSCALE)
    
    # Apply gaussian filter (kernel 5x5)
    image_Gaussian = cv2.GaussianBlur(image, (5, 5), 0)
    
    # Save the modified image (overwrite the original)
    cv2.imwrite(image_path, image_Gaussian)