Upload folder using huggingface_hub

.env

@@ -1,2 +1,3 @@
 api_key = 'eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiJnb29nbGUtb2F1dGgyfDExNjY1NDE1MzQ0MDY1NjEzNTI5MSIsImVtYWlsIjoic2hhaHphaW5oYWlkZXJuYXF2aUBnbWFpbC5jb20iLCJlbWFpbF92ZXJpZmllZCI6dHJ1ZSwiaWF0IjoxNjg1MDAxMjgzMDExfQ.Fb5ODO7KUchlLnrK0KBvSR4pkfIAfYiECRVWXj44RTQ'
-queue_id = 'c0f9e6f8-73d0-42f9-bd4f-700bdf002c04'
+id_key = 'c0f9e6f8-73d0-42f9-bd4f-700bdf002c04'
+passport_key = 'daf046a8-109a-4d2d-a301-0ca90256b0db'

app/app.py

@@ -22,8 +22,9 @@ async def ocr(Id_card: UploadFile = File(...)):
 
     with open(file_path, "wb+") as file_object:
         file_object.write(Id_card.file.read())
-    
-    dictionary = custom_ocr(file_path)
+
+    card_type = "passport"
+    dictionary = custom_ocr(file_path, card_type)
 
     if '' in list(dictionary.values()):
         print("Missing value found in Dic")

app/layoutLM_api/api.py

@@ -7,48 +7,51 @@ from dotenv import load_dotenv
 mimetypes.init()
 load_dotenv()
 
-api_key = os.getenv('api_key')
-queue_id = os.getenv('queue_id')
+api_key =       os.getenv('api_key')
+id_key =        os.getenv('id_key')
+passport_key =  os.getenv('passport_key')
 
+def custom_ocr(image_path, card_type):
 
-def custom_ocr(image_path):
-    
     extracted_field = {
-        'name':             '',
-        'dob':              '',
-        'country':          '',
-        'gender':           '',
-        'document_number':  '',
+        'name': '',
+        'dob': '',
+        'country': '',
+        'gender': '',
+        'document_number': '',
     }
 
-    # PUT THIS IN ENV FILE
-    # Make sure to add your API Key to the auth headers
-    
+    if card_type == "passport":
+        response = Client(api_key).extract_document(passport_key, image_path)
+    elif card_type == "id_card":
+        response = Client(api_key).extract_document(id_key, image_path)
 
-    # Response is a strongly typed object
-    response = Client(api_key).extract_document(queue_id, image_path)
     dictionary = response.to_dict()
-    print("dictionary : ", dictionary)
+    print("dictionary: ", dictionary)
+
+    name_fields = {'Last Name', 'First Name', 'Middle Name'}
+    field_mapping = {
+        'Document Number': 'document_number',
+        'Date of Birth': 'dob',
+        'Birth Date': 'dob',
+        'State': 'country',
+        'County': 'country',
+        'Place of Birth': 'country',
+        'Nationality': 'country',
+        'Sex': 'gender'
+    }
 
     for field in dictionary['formFields']:
+        field_name = field['fieldName']
 
-        if ('Name' in field['fieldName']) and (extracted_field['name'] == ''):
-            
-            if field['fieldName'] == 'Last Name':
-                extracted_field['name'] = field['value']
-            elif field['fieldName'] == 'First Name':
+        if 'Name' in field_name and extracted_field['name'] == '':
+            if field_name in name_fields:
                 extracted_field['name'] = field['value']
-            elif field['fieldName'] == 'Middle Name':
-                extracted_field['name'] = field['value']
-        
-        if field['fieldName'] == 'Document Number':
-            extracted_field['document_number'] = field['value']
-
-        if field['fieldName'] == 'Date of Birth':
-            extracted_field['dob'] = field['value']
 
-        if (field['fieldName'] == 'State' or field['fieldName'] == 'County' or field['fieldName'] == 'Place of Birth' ) and (len(field['value']) > 0):
-            extracted_field['country'] = field['value']
+        if field_name in field_mapping:
+            field_key = field_mapping[field_name]
+            if len(field['value']) > 0:
+                extracted_field[field_key] = field['value']
 
-    print("MODEL EXTRACTED FIELDS : ", extracted_field)
+    print("MODEL EXTRACTED FIELDS: ", extracted_field)
     return extracted_field

deploy.py

@@ -8,7 +8,7 @@ from paddleocr import PaddleOCR
 from app.layoutLM_api.api import custom_ocr
 import cv2
 
-def ocr(image_input, webcam_input):
+def ocr(image_input, webcam_input, card_type_input):
 
     file_path = 'app/images/idcards/input.jpg'
 
@@ -29,8 +29,8 @@ def ocr(image_input, webcam_input):
     
     # with open(file_path, "wb+") as file_object:
     #     file_object.write(input_img.file.read())
-    
-    dictionary = custom_ocr(file_path)
+
+    dictionary = custom_ocr(file_path, card_type_input)
 
     if '' in list(dictionary.values()):
         print("Missing value found in Dic")
@@ -72,7 +72,8 @@ def ocr(image_input, webcam_input):
 # Define the input objects
 image_input = gr.inputs.Image(label="Upload Image")
 webcam_input = gr.inputs.Image(label="Webcam", source="webcam")
+card_type_input = gr.inputs.Dropdown(["passport", "id_card"], label="Card Type")
 
 # Create the Gradio interface
-interface = gr.Interface(fn=ocr, inputs=[image_input, webcam_input], outputs="json")
+interface = gr.Interface(fn=ocr, inputs=[image_input, webcam_input, card_type_input], outputs="json")
 interface.launch()