From 443163864bab56930c2ef735c0aaafddd2561ead Mon Sep 17 00:00:00 2001
From: Matthias Bisping <matthias.bisping@iqser.com>
Date: Sat, 5 Feb 2022 17:59:03 +0100
Subject: [PATCH] implememted clean solution for parsing open tables. still
 needs final refactoring.

---
 scripts/annotate.py       |  2 +-
 vidocp/table_parsing_2.py | 74 +++++++++++++++++++++++++++++++++++++++
 2 files changed, 75 insertions(+), 1 deletion(-)
 create mode 100644 vidocp/table_parsing_2.py

diff --git a/scripts/annotate.py b/scripts/annotate.py
index 4c6d7b8..95de313 100644
--- a/scripts/annotate.py
+++ b/scripts/annotate.py
@@ -1,6 +1,6 @@
 import argparse
 
-from vidocp.table_parsig import annotate_tables_in_pdf
+from vidocp.table_parsing_2 import annotate_tables_in_pdf
 from vidocp.redaction_detection import annotate_boxes_in_pdf
 from vidocp.layout_detection import annotate_layout_in_pdf
 
diff --git a/vidocp/table_parsing_2.py b/vidocp/table_parsing_2.py
new file mode 100644
index 0000000..8b035bf
--- /dev/null
+++ b/vidocp/table_parsing_2.py
@@ -0,0 +1,74 @@
+import cv2
+import matplotlib.pyplot as plt
+import numpy as np
+from pdf2image import pdf2image
+
+
+def add_external_contours(image, img):
+
+    contours, hierarchy = cv2.findContours(img, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)
+
+    for cnt in contours:
+        x, y, w, h = cv2.boundingRect(cnt)
+        cv2.rectangle(image, (x, y), (x + w, y + h), 255, 1)
+
+    return image
+
+
+def isolate_vertical_and_horizontal_components(img_bin):
+
+    line_min_width = 30
+    kernel_h = np.ones((1, line_min_width), np.uint8)
+    kernel_v = np.ones((line_min_width, 1), np.uint8)
+
+    img_bin_h = cv2.morphologyEx(img_bin, cv2.MORPH_OPEN, kernel_h)
+    img_bin_v = cv2.morphologyEx(img_bin, cv2.MORPH_OPEN, kernel_v)
+
+    img_bin_final = img_bin_h | img_bin_v
+
+    return img_bin_final
+
+
+def annotate_image(image, stats):
+
+    image = image.copy()
+
+    for x, y, w, h, area in stats[2:]:
+        if w > 10 and h > 10:
+            cv2.rectangle(image, (x, y), (x + w, y + h), (255, 0, 255), 2)
+
+            for i, (s, v) in enumerate(zip(["x", "y", "w", "h"], [x, y, w, h])):
+                anno = f"{s} = {v}"
+                xann = int(x + 5)
+                yann = int(y + h - (20 * (i + 1)))
+                cv2.putText(image, anno, (xann, yann), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (255, 0, 255), 2)
+
+    return image
+
+
+def parse_table(image: np.array):
+
+    gray_scale = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
+    th1, img_bin = cv2.threshold(gray_scale, 150, 255, cv2.THRESH_BINARY)
+    img_bin = ~img_bin
+
+    img_bin = isolate_vertical_and_horizontal_components(img_bin)
+    img_bin_final = add_external_contours(img_bin, img_bin)
+
+    _, labels, stats, _ = cv2.connectedComponentsWithStats(~img_bin_final, connectivity=8, ltype=cv2.CV_32S)
+
+    return stats
+
+
+def annotate_tables_in_pdf(pdf_path, page_index=1):
+
+    page = pdf2image.convert_from_path(pdf_path, first_page=page_index + 1, last_page=page_index + 1)[0]
+    page = np.array(page)
+
+    stats = parse_table(page)
+    page = annotate_image(page, stats)
+
+    fig, ax = plt.subplots(1, 1)
+    fig.set_size_inches(20, 20)
+    ax.imshow(page)
+    plt.show()