Revert "Update to v2.0"

This reverts commit a06fcbe.

Author: elisemercury

__init__.py

@@ -0,0 +1,145 @@
+import skimage.measure
+import matplotlib.pyplot as plt
+import numpy as np
+import cv2
+import os
+import imghdr
+
+""" 
+Duplicate Image Finder (DIF): function that searches a given directory for images and finds duplicate/similar images among them.
+Outputs the number of found duplicate/similar image pairs with a list of the filenames having lower resolution.
+"""
+
+def compare_images(directory, show_imgs=True, similarity="high", compression=50):
+    """
+    directory (str).........folder to search for duplicate/similar images
+    show_imgs (bool)........True = shows the duplicate/similar images found in output
+                            False = doesn't show found images
+    similarity (str)........"high" = searches for duplicate images, more precise
+                            "low" = finds similar images
+    compression (int).......recommended not to change default value
+                            compression in px (height x width) of the images before being compared
+                            the higher the compression i.e. the higher the pixel size, the more computational ressources and time required                 
+    """
+    # list where the found duplicate/similar images are stored
+    duplicates = []
+    lower_res = []
+    
+    imgs_matrix = create_imgs_matrix(directory, compression)
+
+    # search for similar images
+    if similarity == "low":
+        ref = 1000
+    # search for 1:1 duplicate images
+    else:
+        ref = 200
+
+    main_img = 0
+    compared_img = 1
+    nrows, ncols = compression, compression
+    srow_A = 0
+    erow_A = nrows
+    srow_B = erow_A
+    erow_B = srow_B + nrows       
+    
+    while erow_B <= imgs_matrix.shape[0]:
+        while compared_img < (len(image_files)):
+            # select two images from imgs_matrix
+            imgA = imgs_matrix[srow_A : erow_A, # rows
+                               0      : ncols]  # columns
+            imgB = imgs_matrix[srow_B : erow_B, # rows
+                               0      : ncols]  # columns
+            # compare the images
+            rotations = 0
+            while image_files[main_img] not in duplicates and rotations <= 3:
+                if rotations != 0:
+                    imgB = rotate_img(imgB)
+                err = mse(imgA, imgB)
+                if err < ref:
+                    if show_imgs == True:
+                        show_img_figs(imgA, imgB, err)
+                        show_file_info(compared_img, main_img)
+                    add_to_list(image_files[main_img], duplicates)
+                    check_img_quality(directory, image_files[main_img], image_files[compared_img], lower_res)
+                rotations += 1
+            srow_B += nrows
+            erow_B += nrows
+            compared_img += 1
+        
+        srow_A += nrows
+        erow_A += nrows
+        srow_B = erow_A
+        erow_B = srow_B + nrows
+        main_img += 1
+        compared_img = main_img + 1
+
+    msg = "\n***\n DONE: found " + str(len(duplicates))  + " duplicate image pairs in " + str(len(image_files)) + " total images.\n The following files have lower resolution:"
+    print(msg)
+    return set(lower_res)
+
+# Function that searches the folder for image files, converts them to a matrix
+def create_imgs_matrix(directory, compression):
+    global image_files   
+    image_files = []
+    # create list of all files in directory     
+    folder_files = [filename for filename in os.listdir(directory)]  
+    
+    # create images matrix   
+    counter = 0
+    for filename in folder_files: 
+        if not os.path.isdir(directory + filename) and imghdr.what(directory + filename):
+            img = cv2.imdecode(np.fromfile(directory + filename, dtype=np.uint8), cv2.IMREAD_UNCHANGED)
+            if type(img) == np.ndarray:
+                img = img[...,0:3]
+                img = cv2.resize(img, dsize=(compression, compression), interpolation=cv2.INTER_CUBIC)
+                if counter == 0:
+                    imgs_matrix = img
+                    image_files.append(filename)
+                    counter += 1
+                else:
+                    imgs_matrix = np.concatenate((imgs_matrix, img))
+                    image_files.append(filename)
+    return imgs_matrix
+
+# Function that calulates the mean squared error (mse) between two image matrices
+def mse(imageA, imageB):
+    err = np.sum((imageA.astype("float") - imageB.astype("float")) ** 2)
+    err /= float(imageA.shape[0] * imageA.shape[1])
+    return err
+
+# Function that plots two compared image files and their mse
+def show_img_figs(imageA, imageB, err):
+    fig = plt.figure()
+    plt.suptitle("MSE: %.2f" % (err))
+    # plot first image
+    ax = fig.add_subplot(1, 2, 1)
+    plt.imshow(imageA, cmap = plt.cm.gray)
+    plt.axis("off")
+    # plot second image
+    ax = fig.add_subplot(1, 2, 2)
+    plt.imshow(imageB, cmap = plt.cm.gray)
+    plt.axis("off")
+    # show the images
+    plt.show()
+
+#Function for rotating an image matrix by a 90 degree angle
+def rotate_img(image):
+    image = np.rot90(image, k=1, axes=(0, 1))
+    return image
+
+# Function for printing filename info of plotted image files
+def show_file_info(compared_img, main_img):
+    print("Duplicate file: " + image_files[main_img] + " and " + image_files[compared_img])
+
+# Function for appending items to a list
+def add_to_list(filename, list):
+    list.append(filename)
+
+# Function for checking the quality of compared images, appends the lower quality image to the list
+def check_img_quality(directory, imageA, imageB, list):
+    size_imgA = os.stat(directory + imageA).st_size
+    size_imgB = os.stat(directory + imageB).st_size
+    if size_imgA > size_imgB:
+        add_to_list(imageB, list)
+    else:
+        add_to_list(imageA, list)