Update app.py

app.py

@@ -4,30 +4,39 @@ import pytesseract
 import re
 
 def extract_kyc_fields(image):
-    text = pytesseract.image_to_string(image)
-
-    result = {}
-
-    # Aadhaar number pattern
-    aadhaar_match = re.search(r'\d{4}[\s-]?\d{4}[\s-]?\d{4}', text)
-    if aadhaar_match:
-        aadhaar_number = aadhaar_match.group().replace(" ", "-")
-        result["aadhaar_number"] = aadhaar_number
-
-    # Name: first line (assuming top of image is name)
-    lines = text.split('\n')
-    for line in lines:
-        clean_line = line.strip()
-        if clean_line and clean_line.isalpha() and len(clean_line.split()) >= 2:
-            result["name"] = clean_line
-            break
-
-    # DOB or Birth Date
-    dob_match = re.search(r'(\d{2}[/-]\d{2}[/-]\d{4})|(\d{4}-\d{2}-\d{2})', text)
-    if dob_match:
-        result["dob"] = dob_match.group()
-
-    return result if result else {"error": "Unable to extract KYC details"}
+    try:
+        text = pytesseract.image_to_string(image)
+
+        result = {}
+
+        # Aadhaar number pattern
+        aadhaar_match = re.search(r'\d{4}[\s-]?\d{4}[\s-]?\d{4}', text)
+        if aadhaar_match:
+            aadhaar_number = aadhaar_match.group().replace(" ", "-")
+            result["aadhaar_number"] = aadhaar_number
+        else:
+            result["aadhaar_number"] = "Not found"
+
+        # Name (first all-alphabetic line with 2+ words)
+        lines = text.split('\n')
+        for line in lines:
+            clean_line = line.strip()
+            if clean_line and re.match(r'^[A-Za-z ]+$', clean_line) and len(clean_line.split()) >= 2:
+                result["name"] = clean_line
+                break
+        if "name" not in result:
+            result["name"] = "Not found"
+
+        # DOB or Birth Date
+        dob_match = re.search(r'(\d{2}[/-]\d{2}[/-]\d{4})|(\d{4}-\d{2}-\d{2})', text)
+        if dob_match:
+            result["dob"] = dob_match.group()
+        else:
+            result["dob"] = "Not found"
+
+        return result
+    except Exception as e:
+        return {"error": str(e)}
 
 # Gradio UI
 iface = gr.Interface(
@@ -35,7 +44,7 @@ iface = gr.Interface(
     inputs=gr.Image(type="pil"),
     outputs="json",
     title="🧠 Smart KYC OCR Extractor",
-    description="Upload an Aadhaar/PAN image to extract structured KYC fields using OCR"
+    description="Upload Aadhaar/PAN image and extract KYC fields using Tesseract"
 )
 
 if __name__ == "__main__":