test1

+from paddleocr import PaddleOCR
+from pdf2image import convert_from_path
+import os
+import time
+import numpy as np
+import json
+from pathlib import Path
+import cv2
+from table_detector import detect_tables
+
+# ==== Config ====
+BASE_DIR = os.path.abspath(os.path.join(os.path.dirname(__file__), "..", "..", ".."))
+PDF_NAME = 'aaaa'
+
+# PDF path
+pdf_path = Path(BASE_DIR) / "storage" / "pdf" / "fax.pdf"
+# Output folder
+output_folder = Path(BASE_DIR) / "public" / "image"
+
+#PDF_NAME = pdf_path.stem  # Get the stem of the PDF file
+#print(PDF_NAME)
+
+os.makedirs(output_folder, exist_ok=True)
+
+timestamp = int(time.time())
+img_base_name = f"{PDF_NAME}_{timestamp}"
+
+# ==== OCR Init ====
+ocr = PaddleOCR(
+    use_doc_orientation_classify=False,
+    use_doc_unwarping=False,
+    use_textline_orientation=False
+)
+
+# ==== PDF to Image ====
+pages = convert_from_path(pdf_path, first_page=1, last_page=1)
+image_path = os.path.join(output_folder, f"{img_base_name}.jpg")
+pages[0].save(image_path, "JPEG")
+
+# ==== Run OCR ====
+image_np = np.array(pages[0])
+results = ocr.predict(image_np)
+
+# ==== Convert polygon to bbox ====
+def poly_to_bbox(poly):
+    xs = [p[0] for p in poly]
+    ys = [p[1] for p in poly]
+    return [int(min(xs)), int(min(ys)), int(max(xs)), int(max(ys))]
+
+# ==== Build ocrData ====
+ocr_data_list = []
+for res in results:
+    for text, poly in zip(res['rec_texts'], res['rec_polys']):
+        bbox = poly_to_bbox(poly)
+        ocr_data_list.append({
+            "text": text,
+            "bbox": bbox,
+            "field": "",
+            "hideBorder": False
+        })
+
+# ==== Detect table ====
+table_info = detect_tables(image_path)
+
+# ==== Build JSON ====
+final_json = {
+    "ocr_data": ocr_data_list,
+    "tables": table_info
+}
+
+
+# ==== Save JSON ====
+json_path = os.path.join(output_folder, f"{PDF_NAME}_{timestamp}_with_table.json")
+with open(json_path, "w", encoding="utf-8") as f:
+    json.dump(final_json, f, ensure_ascii=False, indent=2)
+
+print(f"Saved OCR + Table JSON to: {json_path}")

+import cv2
+import numpy as np
+import os
+
+def detect_tables(image_path):
+    img = cv2.imread(image_path)
+    if img is None:
+        raise FileNotFoundError(f"Không đọc được ảnh: {image_path}")
+
+    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
+    blur = cv2.GaussianBlur(gray, (3, 3), 0)
+
+    # Edge detection
+    edges = cv2.Canny(blur, 50, 150, apertureSize=3)
+
+    # --- Horizontal lines ---
+    lines_h = cv2.HoughLinesP(edges, 1, np.pi/180, threshold=120,
+                              minLineLength=int(img.shape[1] * 0.6), maxLineGap=20)
+    ys_candidates, line_segments = [], []
+    if lines_h is not None:
+        for l in lines_h:
+            x1, y1, x2, y2 = l[0]
+            if abs(y1 - y2) <= 3:  # ngang
+                y_mid = int(round((y1 + y2) / 2))
+                ys_candidates.append(y_mid)
+                line_segments.append((x1, x2, y_mid))
+
+    # gom nhóm các y
+    ys, tol_y = [], 10
+    for y in sorted(ys_candidates):
+        if not ys or abs(y - ys[-1]) > tol_y:
+            ys.append(y)
+
+    total_rows = max(0, len(ys) - 1)
+
+    # --- Vertical lines ---
+    lines_v = cv2.HoughLinesP(edges, 1, np.pi/180, threshold=100,
+                              minLineLength=int(img.shape[0] * 0.5), maxLineGap=20)
+    xs = []
+    if lines_v is not None:
+        for l in lines_v:
+            x1, y1, x2, y2 = l[0]
+            if abs(x1 - x2) <= 3:
+                xs.append(int(round((x1 + x2) / 2)))
+
+    # gom nhóm cột
+    x_pos, tol_v = [], 10
+    for v in sorted(xs):
+        if not x_pos or v - x_pos[-1] > tol_v:
+            x_pos.append(v)
+
+    total_cols = max(0, len(x_pos) - 1)
+
+    tables = []
+    if len(ys) >= 3 and line_segments:
+        y_min, y_max = ys[0], ys[-1]
+        min_x = min(seg[0] for seg in line_segments)
+        max_x = max(seg[1] for seg in line_segments)
+        table_box = (min_x, y_min, max_x, y_max)
+
+        rows = []
+        for i in range(len(ys) - 1):
+            row_box = (min_x, ys[i], max_x, ys[i+1])
+            rows.append({"row": tuple(int(v) for v in row_box)})
+            cv2.rectangle(img, (row_box[0], row_box[1]), (row_box[2], row_box[3]), (0, 255, 255), 2)
+
+        tables.append({
+            "total_rows": int(total_rows),
+            "total_cols": int(total_cols),
+            "table_box": tuple(int(v) for v in table_box),
+            "rows_box": rows
+        })
+
+        cv2.rectangle(img, (min_x, y_min), (max_x, y_max), (255, 0, 0), 3)
+
+    debug_path = os.path.splitext(image_path)[0] + "_debug.jpg"
+    cv2.imwrite(debug_path, img)
+
+    return tables