Upload ndarray(image in OpenCV) on to google cloud

Question

I have a similar issues like How to upload a bytes image on Google Cloud Storage from a Python script.

I tried this

from google.cloud import storage
import cv2
from tempfile import TemporaryFile
import google.auth
credentials, project = google.auth.default()
client = storage.Client()
# https://console.cloud.google.com/storage/browser/[bucket-id]/
bucket = client.get_bucket('document')
# Then do other things...
image=cv2.imread('/Users/santhoshdc/Documents/Realtest/15.jpg')
with TemporaryFile() as gcs_image:
    image.tofile(gcs_image)
    blob = bucket.get_blob(gcs_image)
    print(blob.download_as_string())
    blob.upload_from_string('New contents!')
    blob2 = bucket.blob('document/operations/15.png')

    blob2.upload_from_filename(filename='gcs_image')

This is the error that's posing up

> Traceback (most recent call last):   File
> "/Users/santhoshdc/Documents/ImageShapeSize/imageGcloudStorageUpload.py",
> line 13, in <module>
>     blob = bucket.get_blob(gcs_image)   File "/Users/santhoshdc/.virtualenvs/test/lib/python3.6/site-packages/google/cloud/storage/bucket.py",
> line 388, in get_blob
>     **kwargs)   File "/Users/santhoshdc/.virtualenvs/test/lib/python3.6/site-packages/google/cloud/storage/blob.py",
> line 151, in __init__
>     name = _bytes_to_unicode(name)   File "/Users/santhoshdc/.virtualenvs/test/lib/python3.6/site-packages/google/cloud/_helpers.py",
> line 377, in _bytes_to_unicode
>     raise ValueError('%r could not be converted to unicode' % (value,)) ValueError: <_io.BufferedRandom name=7> could not be
> converted to unicode

Can anyone guide me what's going wrong or what I'm doing incorrectly?

Answer 1

As suggested by @A.Queue in(gets deleted after 29 days)

from google.cloud import storage
import cv2
from tempfile import TemporaryFile

client = storage.Client()

bucket = client.get_bucket('test-bucket')
image=cv2.imread('example.jpg')
with TemporaryFile() as gcs_image:
    image.tofile(gcs_image)
    gcs_image.seek(0)
    blob = bucket.blob('example.jpg')
    blob.upload_from_file(gcs_image)

The file got uploaded,but uploading a numpy ndarray doesn't get saved as an image file on the google-cloud-storage

PS:

numpy array has to be convert into any image format before saving.

This is fairly simple, use the tempfile created to store the image, here's the code.

with NamedTemporaryFile() as temp:

    #Extract name to the temp file
    iName = "".join([str(temp.name),".jpg"])

    #Save image to temp file
    cv2.imwrite(iName,duplicate_image)

    #Storing the image temp file inside the bucket
    blob = bucket.blob('ImageTest/Example1.jpg')
    blob.upload_from_filename(iName,content_type='image/jpeg')

    #Get the public_url of the saved image 
    url = blob.public_url

Answer 2

You are calling blob = bucket.get_blob(gcs_image) which makes no sense. get_blob() is supposed to get a string argument, namely the name of the blob you want to get. A name. But you pass a file object.

I propose this code:

with TemporaryFile() as gcs_image:
    image.tofile(gcs_image)
    gcs_image.seek(0)
    blob = bucket.blob('documentation-screenshots/operations/15.png')
    blob.upload_from_file(gcs_image)