First, you must install the Google Cloud SDK, set up and certify your project.
-
Install the Google Cloud SDK: https://cloud.google.com/sdk/docs/install
-
Set up a project with the gcloud init command: https://cloud.google.com/sdk/gcloud/reference/init
-
Create service account key: https://cloud.google.com/docs/authentication/getting-started
-
To grant access to a project, grant the service account the Project > Owner role.
-
To grant a role, find the Select Role list and choose Project > Owner.
- Trap that is translated and not found in searches.
- Trap that is translated and not found in searches.
-
-
Set the path to the environment variable GOOGLE_APPLICATION_CREDENTIALS:.
The following code can perform OCR using the Google Cloud Vision API.
python
import os
from google.cloud import vision
from PIL import Image
os.environ["GOOGLE_APPLICATION_CREDENTIALS"] = "/path/to/your/service-account-key.json"
def ocr_image(image_path):
client = vision.ImageAnnotatorClient()
with open(image_path, 'rb') as image_file:
content = image_file.read()
image = vision.Image(content=content)
response = client.text_detection(image=image)
texts = response.text_annotations
if texts:
return texts[0].description
else:
return ""
def ocr_readable_images(images_dir):
ocr_texts = []
for image_name in sorted(os.listdir(images_dir)):
if image_name.startswith("readable_page"):
image_path = os.path.join(images_dir, image_name)
ocr_text = ocr_image(image_path)
ocr_texts.append(ocr_text)
return ocr_texts
if __name__ == '__main__':
images_dir = 'output/images'
ocr_texts = ocr_readable_images(images_dir)
for i, text in enumerate(ocr_texts):
print(f'Page {i+1}:\n{text}\n{"="*40}')
- It usually takes a little over a second per request.
- Rate limit is 1800 per minute, so I could do about 30 parallels.
Google Cloud Vision, Cloud Vision
This page is auto-translated from [/nishio/Google Cloud Vision](https://scrapbox.io/nishio/Google Cloud Vision) using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.