OCR not able to detect accurately with my photo in date & time

Kah_Jun · June 25, 2025, 8:55am

These are my photos are ready go to detect and arrange them as ascending order with date and time

And here`s my python script:-
import os
import requests
from datetime import datetime

Your OCR.space API Key

API_KEY = ‘K8xxxxxxxxxx’ # Replace with your API Key from OCR.space

Function to call OCR.space API

def extract_date_from_image(image_path):
# OCR.space API URL
ocr_url = “https://api.ocr.space/parse/image”

# Open image file and prepare for API call
with open(image_path, 'rb') as image_file:
    payload = {
        'apikey': API_KEY,
        'language': 'eng',  # Change language if needed (e.g., 'eng' for English)
    }
    files = {
        'file': image_file,
    }
    
    # Call OCR.space API
    response = requests.post(ocr_url, data=payload, files=files)
    
    # Parse the JSON response
    result = response.json()

    # Extract text from the response
    if result['OCRExitCode'] == 1:
        extracted_text = result['ParsedResults'][0]['ParsedText']
        # Print the extracted text to check if the date is included
        print("Extracted Text:", extracted_text)
        return extracted_text
    else:
        print("Error in OCR:", result['ErrorMessage'])
        return None

Function to extract date from the OCR result (basic example, adjust based on date format)

def extract_date(text):
try:
# Example: Detect dates in the format ‘YYYY-MM-DD HH:MM:SS’
detected_date = datetime.strptime(text.split()[0], “%Y-%m-%d”)
return detected_date
except Exception as e:
print(“Error in date extraction:”, e)
return None

Directory containing images

folder_path = r"D:\Downloads\New folder (3)"

Get all image files from the folder

files = [f for f in os.listdir(folder_path) if f.endswith((‘.jpg’, ‘.jpeg’, ‘.png’))]

Extract the dates and sort the images

images_with_dates =
for file in files:
file_path = os.path.join(folder_path, file)
extracted_text = extract_date_from_image(file_path)

if extracted_text:
    detected_date = extract_date(extracted_text)
    if detected_date:
        images_with_dates.append((file, detected_date))

Sort images by the detected date

images_with_dates.sort(key=lambda x: x[1])

Rename files sequentially (1.jpeg, 2.jpeg, 3.jpeg)

for i, (file, _) in enumerate(images_with_dates, start=1):
file_extension = os.path.splitext(file)[1]
new_name = f"{i}{file_extension}"
os.rename(os.path.join(folder_path, file), os.path.join(folder_path, new_name))

print(“Renaming Complete!”)

And become like this when i running in the command prompt
Error in date extraction: time data ‘MAKERZ’ does not match format ‘%Y-%m-%d’
Extracted Text: Go. gle
Juh 25, 2025 11’.48:05 AM
63 Jalan Industri 5
Kampung Baru Sungai Buloh
Sungai Buloh
Selangor

Error in date extraction: time data ‘am’ does not match format ‘%Y-%m-%d’
Extracted Text: MAKERZ O
(GO) Farr-ast
Products Sd
Jun 5;2j25 103985 AM
63 Zalan Industri 5
*kÅmPung Baru Sungai Buloh
Sungaå8uloh
Selangor

How can I fix it to run accurate and detect clearly? Thanks

ocr-api-team · June 27, 2025, 11:43am

Solution: Please switch to OCR Engine 2. It usually works better with text on images.

To test Engine 2 with your images I generated the overlay on our ocr api test page. All looks good and the text is accurately recognized. See here:

So in your code, just change

    payload = {
        'apikey': API_KEY,
        'language': 'eng',  # Change language if needed (e.g., 'eng' for English)
    }

to

    payload = {
        'apikey': API_KEY,
        'ocrengine': 2,
        'language': 'eng',  # Change language if needed (e.g., 'eng' for English)
    }