CollectData.py

import glob
import math
import re
from datetime import datetime
from pathlib import Path

import cv2
import numpy as np
import xlsxwriter

from Algorithms.Minutiae.Libs.matching import match_tuples
from Algorithms.Minutiae.Libs.minutiae import generate_tuple_profile
# Import Minutiae
from Algorithms.Minutiae.Minutiae_OBJ import *
# Import SIFT
from Algorithms.SIFT.SIFT_OBJ import SIFT

#
# # collect Data Here
workbook = xlsxwriter.Workbook(f"Data_Subject_First_600.xlsx")
worksheet = workbook.add_worksheet()
worksheet.set_column(0, 13, 50)
# Set titles here
sheet_titles = {0: "Fingerprint image",

                1: "Alteration Type",

                2: "Match Score (SIFT)",
                3: "Time (SIFT)",
                4: "Verdict (SIFT)",

                5: "Match Score (Minutiae)",
                6: "Time (Minutiae)",
                7: "Verdict (Minutiae)",
                }
for value, title in enumerate(sheet_titles.values()):
    worksheet.write(0, value, title)

# Glob Here
file_pattern = re.compile(r'.*?(\d+).*?')
def get_order(file):
    match = file_pattern.match(Path(file).name)
    if not match:
        return math.inf
    return int(match.groups()[0])

# Path to Real
path_to_real = "C:\\Users\\Ugo\\Desktop\\Fingerprint-Matching-System\\SOCOFing\\Real\\"
# Get Real Images
real_images = []
for img in sorted(glob.glob(f"{path_to_real}*.BMP"),
                  key=get_order):
    real_images.append(img.strip(path_to_real))

# # Get Altered Images
path_to_altered = "C:\\Users\\Ugo\\Desktop\\Fingerprint-Matching-System\\SOCOFing\\Altered\\Altered-Easy\\"
altered_easy = []

try:
    for img in sorted(glob.glob(f"{path_to_altered}*.BMP"), key=get_order):
        # n= cv2.imread(img)
        altered_easy.append(img.strip(path_to_altered))


    counter = 405
    counter_end = counter + 3
    # Initial SIFT
    sift_query = SIFT()
    sift_train = SIFT()
    row= 1
    altered_type ="Easy"
    verdict = ""
    verdict_minutiae = ""
    # Loop to pairs
    for i in range(135,151): # where to end multiply by 6 control where to start and end,
        for j in range(counter, counter_end):
            print(f"Now On {real_images[i]} and {altered_easy[j]}")
            MIN_MATCH_COUNT = 18
            query = cv2.imread(path_to_real+real_images[i],0)
            train = cv2.imread(path_to_altered+altered_easy[j],0)

            start = datetime.now()
            kp1, des1 = sift_query.computeKeypointsAndDescriptors(query)
            kp2, des2 = sift_train.computeKeypointsAndDescriptors(train)
            # # Initialize and use FLANN
            FLANN_INDEX_KDTREE = 1
            index_params = dict(algorithm=FLANN_INDEX_KDTREE, trees=5)
            search_params = dict(checks=37)
            flann = cv2.FlannBasedMatcher(index_params, search_params)
            matches = flann.knnMatch(des1, des2, k=2)
            #
            # # Lowe's ratio test
            good = set()
            for m, n in matches:
                if m.distance < 0.6 * n.distance:
                    good.add(m)
            if len(good) > MIN_MATCH_COUNT:
                src_pts = np.float32([kp1[m.queryIdx].pt for m in good]).reshape(-1, 1, 2)
                dst_pts = np.float32([kp2[m.trainIdx].pt for m in good]).reshape(-1, 1, 2)

                M = cv2.findHomography(src_pts, dst_pts, cv2.RANSAC, 5.0)[0]

            time_taken = datetime.now() - start
        #     Verdict Here
            if len(good) > 35:
                verdict = "Fingerprints/Images Are A Good Match!"
            elif len(good) > 18:
                verdict = "Fingerprints/Images Match With A Low Score!"
            else:
                verdict = "Not enough matches are found"

        #     MINUTIAE
            start = datetime.now()
            coor_termination1, coor_bifurcation1, total_bif_term1 = detectAndComputeMinutiae(path_to_real+real_images[i])
            coor_termination2, coor_bifurcation2, total_bif_term2 = detectAndComputeMinutiae(path_to_altered+altered_easy[j])
            # For caluclation process
            calc_bif_term1 = generate_tuple_profile(total_bif_term1)
            calc_bif_term2 = generate_tuple_profile(total_bif_term2)

            try:
                common_points_both_train, common_points_both_query = match_tuples(calc_bif_term1, calc_bif_term2)
                minutiae_value = len(common_points_both_query)
            except Exception:
                print("Unable to find common points defaulting to score 0")
                minutiae_value = 0
            # # Score here
            # Time ends here
            time_taken_minutiae = datetime.now() - start
        #     Minutiae Verdict
            if minutiae_value >= 7:
                verdict_minutiae = "Fingerprints Are A Good Match"
            elif minutiae_value >= 3:
                verdict_minutiae= "Fingerprints Match With A Really Low Score"
            else:
                verdict_minutiae="Fingerprints do not match"
        #     Record DATA
            # Write Query Image Here
            worksheet.write(row, 0, f"{real_images[i]}\n{altered_easy[j]}")
            # Add Alteration Type
            worksheet.write(row, 1, altered_type)
            #### SIFT ####
            worksheet.write(row, 2, str(len(good)))
            # Time
            worksheet.write(row, 3, time_taken)
            # Verdict
            worksheet.write(row, 4, verdict)
            #--------- MINUTIAE ---------------
            worksheet.write(row,5,str(len(common_points_both_query)))
            worksheet.write(row,6,time_taken_minutiae)
            worksheet.write(row,7,verdict_minutiae)
            row = row + 2
            print("=====================================")
        counter += 3
        counter_end += 3
except Exception:
    print("User stopped operation")
    workbook.close()
print("Saved Workbook successfully")
workbook.close()