idvpackage 3.0.11__py3-none-any.whl → 3.0.12__py3-none-any.whl

idvpackage/uae_id_extraction.py

@@ -1,77 +1,6 @@
-# from pydantic import BaseModel, Field
-# import openai
-# import json
-
-# class UAEIDExtractionResult(BaseModel):
-#     is_header_verified: bool = Field(..., description="Is it the front side of a UAE ID?")
-#     id_number: str = Field(..., description="15-digit UAE ID number")
-
-# def extract_uae_front_id(base64_image: str) -> UAEIDExtractionResult:
-#     """
-#     Extracts UAE ID front fields using OpenAI's vision model and function calling.
-#     Args:
-#         openai_api_key (str): OpenAI API key.
-#         base64_image (str): Base64-encoded image of the UAE ID front.
-#     Returns:
-#         UAEIDExtractionResult: Extracted fields in Pydantic model.
-#     Raises:
-#         Exception: If extraction or parsing fails.
-#     """
-
-#     # Define the function schema for OpenAI function calling
-#     function_schema = {
-#         "name": "UAEIDExtractionResult",
-#         "description": "Extracts fields from the front side of a UAE ID card.",
-#         "parameters": {
-#             "type": "object",
-#             "properties": {
-#                 "is_header_verified": {
-#                     "type": "boolean",
-#                     "description": "Is it the front side of a UAE ID?"
-#                 },
-#                 "id_number": {
-#                     "type": "string",
-#                     "description": "15-digit UAE ID number"
-#                 }
-#             },
-#             "required": ["is_header_verified", "id_number"]
-#         }
-#     }
-#     prompt = (
-#         "You are an expert at extracting information from UAE ID cards. "
-#         "Given an image of the front side of a UAE ID, extract the relevant fields. "
-#         "If the id_number is not found, set it to an empty string. "
-#         "Set is_header_verified to true if the image is the front side of a UAE ID, else false."
-#     )
-#     try:
-#         response = openai.ChatCompletion.create(
-#             model="gpt-4o",
-#             messages=[
-#                 {"role": "system", "content": "You are a helpful assistant."},
-#                 {"role": "user", "content": [
-#                     {"type": "text", "text": prompt},
-#                     {"type": "image_url", "image_url": {"url": f"data:image/jpeg;base64,{base64_image}"}}
-#                 ]}
-#             ],
-#             functions=[function_schema],
-#             function_call={"name": "UAEIDExtractionResult"},
-#             max_tokens=300
-#         )
-#         message = response.choices[0].message
-#         if message.function_call and message.function_call.arguments:
-#             args = json.loads(message.function_call.arguments)
-#             return UAEIDExtractionResult(**args)
-#         else:
-#             return {'error':'covered_photo'}
-#     except Exception as e:
-#         return {'error':'covered_photo'}
-
-
-
-import base64
+
 import time
 from io import BytesIO
-from typing import Optional
 import cv2
 
 from openai import OpenAI
@@ -101,8 +30,16 @@ Extract ALL fields from this UAE National ID **front side** image with high accu
 
 Return a JSON object with the following fields (use the exact field names):
 
-- id_number: The ID number exactly as shown on the card (example: 784-1999-1234567-1 but return without dashes, e.g., 784199912345671)
-- header_verified: Return True if one of the texts present in the image is "UNITED ARAB EMIRATES"; otherwise False.
+- id_number: The ID number exactly as shown on the card example: 784-1999-1234567-1 but return without dashes (i.e., 784199912345671)
+- name: Full name as shown on the card (extract exactly as written on the card).
+- first_name: Only the First name, as shown on the card (extract exactly as written on the card).
+- last_name: Only the Last name, as shown on the card (extract exactly as written on the card).
+- dob: Extract date of birth exactly as shown on card. Convert date to format DD-MM-YYYY If there is no date of birth, return an empty string.
+- nationality: Extract the nationality exactly as printed on the card Convert nationality to ISO 3166-1 alpha-3 code. e.g. EGY for Egypt, IND for India.
+- issue_date: Extract issue date exactly as shown on card. Convert date to format DD-MM-YYYY If there is no issue date, return an empty string.
+- expiry_date: Extract the expiry date exactly as shown on card. Convert date to format DD-MM-YYYY If there is no expiry date, return an empty string.
+- gender: Extract gender exactly as printed on the card. If there is no gender, return an empty string. If gender is M, return "Male". If gender is F, return "Female".
+- header_verified: Return True if one of the texts present in the image UNITED ARAB EMIRATES else return False.
 
 Instructions:
 - Do NOT guess or hallucinate any values. If unclear, return empty string.
@@ -116,62 +53,98 @@ Extract ALL fields from this UAE National ID **back side** image with high accur
 
 Return a JSON object with the following fields (use the exact field names):
 
-- gender_back:  If not present as a separate field, return null. Gender as either M or F. Only extract if it is printed as a separate field on the back side of the card. Do NOT extract from the MRZ.
-- date_of_birth_back: If not present as a separate field, return null. Date of birth exactly as shown on the back side of the card preserve DD/MM/YY format. Only extract if it is printed as a separate field. Do NOT extract from the MRZ.
-- date_of_expiry_back: If not present as a separate field, return null. Expiry date exactly as shown on the back side of the card preserve DD/MM/YY format. Only extract if it is printed as a separate field. Do NOT extract from the MRZ.
-- employer: Employer name as shown in the image (English version; extract exactly as written on the card)
-- occupation: Occupation as shown in the image (English version; extract exactly as written on the card)
+- employer: Employer name as shown in the image (English version; extract exactly as written on the card). If the employer is not present, return an empty string.
+- occupation: Occupation as shown in the image (English version; extract exactly as written on the card). If the occupation is not present, return an empty string.
+- family_sponsor : Family sponsor name as show in the image English Version(extract exactly as written on the card) If the family sponsor is not present, return an empty string.
+- expiry_date: Extract the expiry date exactly as shown on card. Convert date to format DD-MM-YYYY If there is no expiry date, return an empty string.
+- dob: Extract date of birth date exactly as shown on card. Convert date to format DD-MM-YYYY If there is no date of birth, return an empty string
+- gender: Extract gender exactly as shown on card. If it is M, return Male If it is F, return Female If there is no gender on the card, return an empty string.
 - card_number: Card number as shown on the card (extract exactly as written, e.g., 119887248)
-- issuing_place: Issuing place as shown in the image (extract exactly as written on the card)
+- issuing_place: Issuing place as shown in the image (extract exactly as written on the card). If the issuing place is not present, return an empty string.
 - back_header_verified: Return True if one of the texts present in the image is "ILARE" or "IDLARE"; otherwise False.
 - mrz1: First line of the MRZ
 - mrz2: Second line of the MRZ
 - mrz3: Third line of the MRZ
-- family_sponsor: Family sponsor name as shown in the image (English version; extract exactly as written on the card)
+- family_sponsor: Family sponsor name as shown in the image (English version; extract exactly as written on the card). If the family sponsor is not present, return an empty string.
+- dob_mrz: Date of birth extracted from MRZ line 2 in DD-MM-YYYY format.
+- gender_mrz: Gender extracted from MRZ line 2: 'M' for Male, 'F' for Female.
+- expiry_date_mrz: Expiry date extracted from MRZ line 2 in DD-MM-YYYY format.
 
 Instructions:
 - Do NOT guess or hallucinate any values. If unclear, return empty string.
 - Only use information visible on the card.
 - Return the result as a single JSON object matching the schema above.
+- If any of the fields are not present on the card, return empty strings for those fields.
 """
 
 class UAEFront(BaseModel):
 
-    id_number: str = Field(...,
+    id_number: str = Field(...,min_length=15, maximum=15,
         description = "The ID number exactly as shown on the card example: 784-1999-1234567-1 but return without dashes (i.e., 784199912345671)",
     )
 
+    name: str = Field(..., description="Full name as shown on the card (extract exactly as written on the card).")
+
+    first_name: str = Field(..., description="First name as shown on the card (extract exactly only the first name as written on the card).")
+
+    last_name: str = Field(..., description="Last name as shown on the card (extract exactly only the last name as written on the card).")
+
+    dob: str = Field(..., description="Extract date of birth exactly as shown on card."
+                                      "Convert date to format dd-mm-yyyy"
+                                      "If there is no date of birth, return an empty string.")
+
+    nationality: str = Field(..., description="Extract the nationality exactly as printed on the card"
+                                              "Convert nationality to ISO 3166-1 alpha-3 code. e.g. EGY for Egypt, IND for India.")
+
+    issue_date: str = Field(..., description="Extract issue date exactly as shown on card."
+                                             "Convert date to format dd-mm-yyyy"
+                                             "If there is no issue date, return an empty string.")
+
+    expiry_date: str = Field(..., description="Extract the expiry date exactly as shown on card."
+                                              "Convert date to format dd-mm-yyyy"
+                                              "If there is no expiry date, return an empty string.")
+
+    gender: str = Field(..., description="Extract gender exactly as printed on the card."
+                                         "If there is no gender, return an empty string.")
+
     
     header_verified: bool = Field(
         ...,
-        description=" Return True if one of the texts present in the image UNITED ARAB EMIRATES",
+        description=" Return True if one of the texts present in the image UNITED ARAB EMIRATES else return False.",
     )
 
 class UAEBack(BaseModel):
 
-    gender_back: str = Field(...,
-        description=" If not present as a separate field, return null. Gender as either M or F. Only extract if it is printed as a separate field on the back side of the card. Do NOT extract from the MRZ."
+    employer: str = Field(...,
+        description = "Employer name as shown in the image English Version(extract exactly as written on the card)"
+                      "If the employer is not present, return an empty string."
     )
 
-    date_of_birth_back: str = Field(...,
-        description = "If not present as a separate field, return null. Date of birth exactly as shown on the back side of the card. Only extract if it is printed as a separate field. Do NOT extract from the MRZ."
+    occupation: str = Field(...,
+        description = "Occupation as shown in the image English version(extract exactly as written on the card)"
+                      "If the occupation is not present, return an empty string."
     )
 
-    date_of_expiry_back: str = Field(...,
-        description = "If not present as a separate field, return null. The expiry date exactly as shown on the back side of the card (preserve original format). Only extract if it is printed as a separate field. Do NOT extract from the MRZ."
+    family_sponsor : str = Field(...,
+        description="Family sponsor name as show in the image English Version(extract exactly as written on the card)"
+                    "If the family sponsor is not present, return an empty string."
     )
-   
 
-    employer: str = Field(...,
-        description = "Employer name as show in the image English Version(extract exactly as written on the card)",
-    )
+    expiry_date: str = Field(..., description="Extract the expiry date exactly as shown on card."
+                                              "Convert date to format dd-mm-yyyy"
+                                              "If there is no expiry date, return an empty string.")
 
-    occupation: str = Field(...,
-        description = "Occupation as show in the image English version(extract exactly as written on the card)",
-    )
+    dob: str = Field(..., description="Extract date of birth date exactly as shown on card."
+                                      "Convert date to format dd-mm-yyyy"
+                                      "If there is no date of birth, return an empty string.")
 
-    card_number: str = Field(...,
-        description = "Card number extract exactly as written on the card ex: 119887248"
+    gender: str = Field(..., description="Extract gender exactly as shown on card."
+                                         "If it is M, return Male"
+                                         "If it is F, return Female"
+                                         "If there is no gender on the card, return an empty string.")
+
+    card_number: str = Field(...,min_length=9, max_length=9,
+        description = "9-digit Card number extract exactly as written on the card, example: 119887248"
     )
 
     issuing_place: str = Field(...,
@@ -180,26 +153,29 @@ class UAEBack(BaseModel):
 
     back_header_verified: bool = Field(
         ...,
-        description=" Return True if one of the texts present in the image ILARE or  IDARE  ",
+        description=" Return True if one of the texts present in the image: ILARE or IDARE else return False  ",
     )
 
     mrz1: str = Field(..., 
-        description="First line of the MRZ"
+        description="First line of the MRZ, exactly as printed on the card."
     )
 
     mrz2: str = Field(..., 
-        description="Second line of the MRZ"
+        description="Second line of the MRZ, exactly as printed on the card."
     )
 
     mrz3: str = Field(..., 
-        description="Third line of the MRZ"
+        description="Third line of the MRZ, exactly as printed on the card."
     )
-    
-    family_sponsor : str = Field(...,
-        description="Family sponsor name as show in the image English Version(extract exactly as written on the card)",
+    dob_mrz: str = Field(...,
+        description="Date of birth extracted from MRZ line 2 in dd-mm-yyyy format."
+    )
+    gender_mrz: str = Field(...,
+        description="Gender extracted from MRZ line 2: 'M' for Male, 'F' for female."       
+    )
+    expiry_date_mrz: str = Field(...,
+        description="Expiry date extracted from MRZ line 2 in dd-mm-yyyy format."
     )
-
-
 
 
 def process_image(side):
@@ -217,8 +193,8 @@ def process_image(side):
 
     return model, prompt
 
-def get_openai_response(prompt: str, model_type, image: BytesIO, genai_key):
-    b64_image = base64.b64encode(image.getvalue()).decode("utf-8")
+def get_openai_response(prompt: str, model_type, image: str, genai_key):
+
     for attempt in range(3):
         try:
             client = OpenAI(api_key=genai_key)
@@ -228,7 +204,7 @@ def get_openai_response(prompt: str, model_type, image: BytesIO, genai_key):
                     {"role": "system", "content": "You are an expert at extracting information from identity documents."},
                     {"role": "user", "content": [
                         {"type": "input_text", "text": prompt},
-                        {"type": "input_image", "image_url": f"data:image/jpeg;base64,{b64_image}", "detail": "low"},
+                        {"type": "input_image", "image_url": f"data:image/jpeg;base64,{image}", "detail": "low"},
                     ]},
                 ],
                 text_format=model_type,
@@ -239,50 +215,10 @@ def get_openai_response(prompt: str, model_type, image: BytesIO, genai_key):
             time.sleep(2)
     return None
 
-def _image_to_jpeg_bytesio(image) -> BytesIO:
-    """
-    Accepts: numpy.ndarray (OpenCV BGR), PIL.Image.Image, bytes/bytearray, or io.BytesIO
-    Returns: io.BytesIO containing JPEG bytes (ready for get_openai_response)
-    """
-    import numpy as np
-
-    if isinstance(image, BytesIO):
-        image.seek(0)
-        return image
-
-    if isinstance(image, (bytes, bytearray)):
-        return BytesIO(image)
-
-    try:
-        from PIL.Image import Image as _PILImage
-
-        if isinstance(image, _PILImage):
-            buf = BytesIO()
-            image.convert("RGB").save(buf, format="JPEG", quality=95)
-            buf.seek(0)
-            return buf
-    except Exception:
-        pass
-
-    if isinstance(image, np.ndarray):
-        success, enc = cv2.imencode(".jpg", image)
-        if not success:
-            raise ValueError("cv2.imencode failed")
-        return BytesIO(enc.tobytes())
-
-    raise TypeError(
-        "Unsupported image type. Provide numpy.ndarray, PIL.Image.Image, bytes, or io.BytesIO."
-    )
-
 def get_response_from_openai_uae(image, side, country, openai_key):
 
-    logging.info("Processing image for UAE passport extraction OPENAI......")
+    logging.info("Processing image for UAE National ID extraction OPENAI......")
     logging.info(f" and type: {type(image)}")
-    try:
-        image = _image_to_jpeg_bytesio(image)
-    except Exception as e:
-        logging.error(f"Error encoding image: {e}")
-        return {"error": "Image encoding failed"}
     try:
         model, prompt = process_image(side)
         logging.info(f"Using model: {model.__name__} and prompt {prompt[:100]}")

{idvpackage-3.0.11.dist-info → idvpackage-3.0.12.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: idvpackage
-Version: 3.0.11
+Version: 3.0.12
 Summary: This repository contains a Python program designed to execute Optical Character Recognition (OCR) and Facial Recognition on images.
 Home-page: https://github.com/NymCard-Payments/project_idv_package
 Classifier: Programming Language :: Python :: 3

idvpackage-3.0.12.dist-info/RECORD

@@ -0,0 +1,34 @@
+idvpackage/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+idvpackage/blur_detection.py,sha256=CtLqTtg2azE4YCM8AT1jYJq6QxGoniMO0VkQYIcMtRI,1883
+idvpackage/common.py,sha256=xhDBZqjIlzhNqIkHzH3Hy7Sdr0ncD9-BdSE091MDbA4,15243
+idvpackage/constants.py,sha256=XvQy5ORGsXoQGcXcvUBC17vpFMzgzleS9S_mDN72G4c,7212
+idvpackage/iraq_id_extraction_withopenai.py,sha256=r72M3osCDqYjctqTuhPvkTxxjIST5ZpBFHVUDPzUqOo,21884
+idvpackage/jor_passport_extraction.py,sha256=O3qzYCbgm_tWzPKJc3b9qrH0VrefCQVPKNJdOHYku84,9214
+idvpackage/lebanon_id_extraction.py,sha256=eqWSpfb83RRk5TEhJNOuiPLt-wCLY5L4TvrpkOckSPs,21192
+idvpackage/liveness_spoofing_v2.py,sha256=G3sCu4LkzhJvL33S75sI_QTpc1nA3wo9SqG1elkaIwg,7409
+idvpackage/ocr.py,sha256=03sIUsWnMsoGp9dQ5eu0VKmPrHxjZuDqSqj107Y-OKk,156980
+idvpackage/ocr_utils.py,sha256=ro028PvFUdZjzlOFqOijyRI2VvAGGAypAzyU_Xp6Ucs,77285
+idvpackage/pse_passport_extraction.py,sha256=fDrvCzN9PWhqYYvULVD89EYeBWrcMazpzY02uSlus-0,8627
+idvpackage/qatar_id_extraction.py,sha256=iTGB7j8b7aHi69ZlH4pekPD9klavqXxOSZvzwSZBL2M,9044
+idvpackage/sudan_passport_extraction.py,sha256=fgmRLB5WdA9RqQOkP_J2lDMcgDbdKh9L9XQaWC5UQPY,17504
+idvpackage/syr_passport_extraction.py,sha256=ga7wSLi3ydDxHo0GNwCf3-XLg5PqcTW40SS7F8XjLN8,9297
+idvpackage/uae_id_extraction.py,sha256=_kp1FgzEoxvMTCzqXJ5Lej3rlkK6lGPKZCPyQWrscoI,11338
+idvpackage/icons/battery1.png,sha256=cnc3nngw9baelwSeihjzZzAfRhhvzXpDJvrqgLeVYIw,2616
+idvpackage/icons/battery3.png,sha256=2hwjcuhBaskFRuTE0i85wT82wvJ9ftm7ALydrfbMp1A,1187
+idvpackage/icons/network1.png,sha256=2W_DhrNn6HfwyIe4VK-jv5MmMkhzvQ_oR1BOD3FYzmA,1303
+idvpackage/icons/network2.png,sha256=0jdYKl4a41EjDAW0AxN-auPxcxB3CGoI6hTrr7iFNx4,1067
+idvpackage/icons/wifi1.png,sha256=NMk17e66G8EZpYi_i0MKEsW81QfqecGaiQx1C6Y7_Uo,1670
+idvpackage/icons/wifi3.png,sha256=RuHLFYhp265UWEUzsAy_V9omVxFYcPt3P8_JvUSddko,1385
+idvpackage/icons/wifi4.png,sha256=jaTfaGlVu5lktb_souVF6OLuSnCsRIvpfP5gBW0951w,2565
+idvpackage/spoof_resources/2.7_80x80_MiniFASNetV2.pth,sha256=pesC4YQ_GbU4a5U8xMnwEcP5hdDuK7mBnuqaFCCZvsA,1849453
+idvpackage/spoof_resources/4_0_0_80x80_MiniFASNetV1SE.pth,sha256=hO4dN9lolNXoLeWlffBE74Cli-KyGLXtfN_YdewvWZA,1856130
+idvpackage/spoof_resources/MiniFASNet.py,sha256=ZbOVkjsLYqsr90Ac8OV-2G7AwwDm2CUVB1H6WhmTFZ4,12345
+idvpackage/spoof_resources/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+idvpackage/spoof_resources/functional.py,sha256=164aGMvBDHU6HWRyDBr-2EUPeJ4rc4cXl10NJyGB7eE,20138
+idvpackage/spoof_resources/generate_patches.py,sha256=ANvJ5rDl-SXARchS8JwOxuKLhUuabbxLwvz4zsYbADE,1722
+idvpackage/spoof_resources/transform.py,sha256=3Ft9S6g6N0SU24f3feHXquh5qRc85JFt2gNKnCYQwjo,11311
+idvpackage-3.0.12.dist-info/licenses/LICENSE,sha256=JTmNGOPPvG2XBgkW2R2xwzJeR_OEjaFKeePN1jGYVt8,1068
+idvpackage-3.0.12.dist-info/METADATA,sha256=21m7s8UBQxFK5Ohy9xeS4vJdiJb9We75qjujFrurzwg,4800
+idvpackage-3.0.12.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+idvpackage-3.0.12.dist-info/top_level.txt,sha256=8jL1PsYvKBIWWLGe8jEm7kU5yZ5sld_OCaz0Dzvi6qY,11
+idvpackage-3.0.12.dist-info/RECORD,,

{idvpackage-3.0.11.dist-info → idvpackage-3.0.12.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (80.9.0)
+Generator: setuptools (80.10.2)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

idvpackage/ekyc.py

@@ -1,78 +0,0 @@
-import requests
-from idvpackage.constants import *
-from idvpackage.ocr_utils import fuzzy_match_fields
-import openai
-
-class Ekyc:
-
-    def __init__(self):
-        self.maps_endpoint = GOOGLE_MAPS_API_ENDPOINT
-        self.gmaps_api_key = API_KEY
-        self.openai_key = OPENAI_API_KEY
-
-    def classify_address(self, address):
-        openai.api_key = self.openai_key
-        prompt = f"Consider yourself as an address guide, your task is to do address classification from the address that you receive. you'll only respond in 1 word that is either residential or commercial. here is an address, tell me if this address is residential or commercial: {address}"
-        
-        response = openai.ChatCompletion.create(
-            model="gpt-3.5-turbo-16k",
-            messages=[
-                {"role": "user", "content": prompt}
-            ],
-            temperature=0.2,
-            max_tokens=1000
-        )
-        return response.choices[0]["message"]["content"]
-
-    def verify_address(self, address):
-        endpoint = self.maps_endpoint
-        
-        params = {
-            "address": address,
-            "key": self.gmaps_api_key,
-        }
-
-        try:
-            response = requests.get(endpoint, params=params)
-            data = response.json()
-            if response.status_code == 200:
-                if data['status'] == 'OK':
-                    for result in data['results']:
-                        for address_type in result['types']:
-    #                         print(result['types'])
-    #                         if address_type in ['street_address', 'subpremise', 'premise']:
-    #                             return "Residential", result['formatted_address'], result['types']
-                            if address_type in ['food', 'restaurant', 'lodging', 'business', 'general_contractor', 'hair_care', 'health', 'spa']:
-                                return True, "Commercial"
-                            
-                        return True, "Residential"
-                    return False, "Unknown"
-                else:
-                    return False, None
-            else:
-                return False, None
-        
-        except Exception as e:
-            return False, None
-        
-    def address_validation(self, user_input_address, utility_bill_address, address_from_other_source):
-        res1 = fuzzy_match_fields(user_input_address, address_from_other_source)
-        res2 = fuzzy_match_fields(user_input_address, utility_bill_address)
-
-        if res1 or res2:
-            return True
-        else:
-            return False
-    
-    def address_verification_and_validation(self, address):
-        address = {
-            'error': ''
-        }
-        result = self.verify_address(address)
-        classification_result = self.classify_address(address)
-
-        if result and classification_result.lower() == 'residential':
-            return address
-        
-        else:
-            address['error'] = 'address_caution'