Added collected data

Mini-Sylar · Oct 20, 2022 · 2ea545e · 2ea545e
1 parent 9f91aeb
commit 2ea545e
Show file tree

Hide file tree

Showing 19 changed files with 101 additions and 99 deletions.
diff --git a/Algorithms/Data/Data_Subject_First_900.xlsx b/Algorithms/Data/Data_Subject_First_900.xlsx
diff --git a/...a_Subject_1_Altered_Easy_Left_Hand_M.xlsx → ...a_Subject_1_Altered_Easy_Left_Hand_M.xlsx b/...a_Subject_1_Altered_Easy_Left_Hand_M.xlsx → ...a_Subject_1_Altered_Easy_Left_Hand_M.xlsx
diff --git a/..._Subject_1_Altered_Easy_Right_Hand_M.xlsx → ..._Subject_1_Altered_Easy_Right_Hand_M.xlsx b/..._Subject_1_Altered_Easy_Right_Hand_M.xlsx → ..._Subject_1_Altered_Easy_Right_Hand_M.xlsx
diff --git a/...Subject_1_Altered_Medium_Left_Hand_M.xlsx → ...Subject_1_Altered_Medium_Left_Hand_M.xlsx b/...Subject_1_Altered_Medium_Left_Hand_M.xlsx → ...Subject_1_Altered_Medium_Left_Hand_M.xlsx
diff --git a/...a_Subject_2_Altered_Easy_Left_Hand_F.xlsx → ...a_Subject_2_Altered_Easy_Left_Hand_F.xlsx b/...a_Subject_2_Altered_Easy_Left_Hand_F.xlsx → ...a_Subject_2_Altered_Easy_Left_Hand_F.xlsx
diff --git a/..._Subject_2_Altered_Easy_Right_Hand_F.xlsx → ..._Subject_2_Altered_Easy_Right_Hand_F.xlsx b/..._Subject_2_Altered_Easy_Right_Hand_F.xlsx → ..._Subject_2_Altered_Easy_Right_Hand_F.xlsx
diff --git a/...a_Subject_3_Altered_Easy_Left_Hand_M.xlsx → ...a_Subject_3_Altered_Easy_Left_Hand_M.xlsx b/...a_Subject_3_Altered_Easy_Left_Hand_M.xlsx → ...a_Subject_3_Altered_Easy_Left_Hand_M.xlsx
diff --git a/..._Subject_3_Altered_Easy_Right_Hand_M.xlsx → ..._Subject_3_Altered_Easy_Right_Hand_M.xlsx b/..._Subject_3_Altered_Easy_Right_Hand_M.xlsx → ..._Subject_3_Altered_Easy_Right_Hand_M.xlsx
diff --git a/...a_Subject_4_Altered_Easy_Left_Hand_M.xlsx → ...a_Subject_4_Altered_Easy_Left_Hand_M.xlsx b/...a_Subject_4_Altered_Easy_Left_Hand_M.xlsx → ...a_Subject_4_Altered_Easy_Left_Hand_M.xlsx
diff --git a/..._Subject_4_Altered_Easy_Right_Hand_M.xlsx → ..._Subject_4_Altered_Easy_Right_Hand_M.xlsx b/..._Subject_4_Altered_Easy_Right_Hand_M.xlsx → ..._Subject_4_Altered_Easy_Right_Hand_M.xlsx
diff --git a/...a_Subject_5_Altered_Easy_Left_Hand_M.xlsx → ...a_Subject_5_Altered_Easy_Left_Hand_M.xlsx b/...a_Subject_5_Altered_Easy_Left_Hand_M.xlsx → ...a_Subject_5_Altered_Easy_Left_Hand_M.xlsx
diff --git a/..._Subject_5_Altered_Easy_Right_Hand_M.xlsx → ..._Subject_5_Altered_Easy_Right_Hand_M.xlsx b/..._Subject_5_Altered_Easy_Right_Hand_M.xlsx → ..._Subject_5_Altered_Easy_Right_Hand_M.xlsx
diff --git a/Algorithms/GUI/CollectData.py b/Algorithms/GUI/CollectData.py
@@ -55,101 +55,105 @@ def get_order(file):
 # # Get Altered Images
 path_to_altered = "C:\\Users\\Ugo\\Desktop\\Fingerprint-Matching-System\\SOCOFing\\Altered\\Altered-Easy\\"
 altered_easy = []
-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 = 0
-counter_end = 3
-# Initial SIFT
-sift_query = SIFT()
-sift_train = SIFT()
-row= 1
-altered_type ="Easy"
-verdict = ""
-verdict_minutiae = ""
-# Loop to pairs
-for i in range(0,100): # 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)
-        kp1, des1 = sift_query.computeKeypointsAndDescriptors(query)
-        kp2, des2 = sift_train.computeKeypointsAndDescriptors(train)
-
-        start = datetime.now()
-
-        # # 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
 
+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()
diff --git a/Algorithms/GUI/Data_Subject.xlsx b/Algorithms/GUI/Data_Subject.xlsx
diff --git a/Algorithms/GUI/Data_Subject2.xlsx b/Algorithms/GUI/Data_Subject2.xlsx
diff --git a/Algorithms/Minutiae/Libs/matching.py b/Algorithms/Minutiae/Libs/matching.py
@@ -41,10 +41,8 @@ def match_tuples(tuple_base: dict, tuple_test: dict, th_range: float = .01, th_a
     # Tuple-wise comparison with all tuple profiles in the test image.
     for i, (m, [ratios, angles]) in enumerate(tuple_base.items()):
         # Explore matching ratios.
-        try:
-            matching_values = (ratios_test == ratios).sum(1)
-        except Exception:
-            matching_values = (ratios_test == ratios).sum(0)
+
+        matching_values = (ratios_test == ratios).sum(1)
         # Tuples found to match with this base tuple. 
         matching_indices = np.where((matching_values == matching_values.max()) & (matching_values >= 2))[0]
 

diff --git a/SOCOFing/Altered/Altered-Easy/7__M_Right_index_finger_Obl.BMP b/SOCOFing/Altered/Altered-Easy/7__M_Right_index_finger_Obl.BMP
diff --git a/SOCOFing/Altered/Altered-Medium/2__F_Left_little_finger_Zcut.BMP b/SOCOFing/Altered/Altered-Medium/2__F_Left_little_finger_Zcut.BMP
diff --git a/SOCOFing/Altered/Altered-Medium/4__M_Right_middle_finger_CR.BMP b/SOCOFing/Altered/Altered-Medium/4__M_Right_middle_finger_CR.BMP