Update app.py

app.py

@@ -2,14 +2,29 @@ import gradio as gr
 from ultralytics import YOLO
 from PIL import Image, ImageDraw
 import pytesseract
+import subprocess
 
-# Set the correct Tesseract path for Hugging Face Spaces
+# Set Tesseract path (Ensure it works on Hugging Face Spaces)
 pytesseract.pytesseract.tesseract_cmd = "/usr/bin/tesseract"
 
+# Load YOLO model
 YOLO_MODEL_PATH = "best.pt"
 model = YOLO(YOLO_MODEL_PATH, task='detect').to("cpu")
 
+def check_tesseract():
+    """Check if Tesseract is installed and print its version."""
+    try:
+        tesseract_path = subprocess.check_output(["which", "tesseract"]).decode("utf-8").strip()
+        tesseract_version = subprocess.check_output(["tesseract", "--version"]).decode("utf-8").split("\n")[0]
+        print(f"Tesseract Path: {tesseract_path}")
+        print(f"Tesseract Version: {tesseract_version}")
+        return True
+    except Exception as e:
+        print(f"Tesseract not found: {e}")
+        return False
+
 def merge_boxes_into_lines(boxes, y_threshold=10):
+    """Merge bounding boxes if they belong to the same text row."""
     if len(boxes) == 0:
         return []
 
@@ -32,6 +47,7 @@ def merge_boxes_into_lines(boxes, y_threshold=10):
     return merged_lines
 
 def detect_and_ocr(image):
+    """Detects text lines, draws bounding boxes, and runs OCR if available."""
     image = Image.fromarray(image)
     original_image = image.copy()
 
@@ -50,17 +66,21 @@ def detect_and_ocr(image):
 
         cropped_line = image.crop((x1, y1, x2, y2))
 
-        ocr_text = pytesseract.image_to_string(cropped_line, lang="eng").strip()
-        if ocr_text:
-            extracted_text_lines.append(ocr_text)
+        if check_tesseract():  # If Tesseract is installed, run OCR
+            try:
+                ocr_text = pytesseract.image_to_string(cropped_line, lang="khm").strip()
+                if ocr_text:
+                    extracted_text_lines.append(ocr_text)
+            except Exception as e:
+                print(f"OCR failed for line {idx}: {e}")
+
+    full_text = "\n".join(extracted_text_lines) if extracted_text_lines else "⚠️ OCR not available. Showing detected lines only."
 
-    full_text = "\n".join(extracted_text_lines)
-    
     return original_image, full_text
 
 with gr.Blocks() as iface:
-    gr.Markdown("# Text Line Detection with OCR")
-    gr.Markdown("## Upload an image to detect text lines and extract text")
+    gr.Markdown("# Text Line Detection with Khmer OCR")
+    gr.Markdown("## Upload an image to detect text lines and extract Khmer text")
 
     with gr.Row():
         with gr.Column(scale=1):