How to annotate MULTIPLE images from a single call using Google's vision API? Python

问题

I recently started using Google's vision API. I am trying to annotate a batch of images and therefore issued the 'batch image annotation offline' guide from their documentation.

However, it is not clear to me how I can annotate MULTIPLE images from one API call. So let's say I have stored 10 images in my google cloud bucket. How can I annotate all these images at once and store them in one JSON file? Right now, I wrote a program that calls their example function and it works, but to put it simple, why can't I say: 'Look in this folder and annotate all images in it.'?

Thanks in advance.

from batch_image_labeling import sample_async_batch_annotate_images
counter = 0
for file in os.listdir('my_directory'):
    filename = file
    sample_async_batch_annotate_images('gs://my_bucket/{}'.format(filename), 'gs://my_bucket/{}'.format(counter))
    counter += 1


from google.cloud import vision_v1
from google.cloud.vision_v1 import enums
import six

def sample_async_batch_annotate_images(input_image_uri, output_uri):
  """Perform async batch image annotation"""

  client = vision_v1.ImageAnnotatorClient()

  if isinstance(input_image_uri, six.binary_type):
    input_image_uri = input_image_uri.decode('utf-8')
  if isinstance(output_uri, six.binary_type):
    output_uri = output_uri.decode('utf-8')
  source = {'image_uri': input_image_uri}
  image = {'source': source}
  type_ = enums.Feature.Type.LABEL_DETECTION
  features_element = {'type': type_}
  type_2 = enums.Feature.Type.IMAGE_PROPERTIES
  features_element_2 = {'type': type_2}
  features = [features_element, features_element_2]
  requests_element = {'image': image, 'features': features}
  requests = [requests_element]
  gcs_destination = {'uri': output_uri}

  # The max number of responses to output in each JSON file
  batch_size = 2
  output_config = {'gcs_destination': gcs_destination, 'batch_size': batch_size}

  operation = client.async_batch_annotate_images(requests, output_config)

  print('Waiting for operation to complete...')
  response = operation.result()

  # The output is written to GCS with the provided output_uri as prefix
  gcs_output_uri = response.output_config.gcs_destination.uri
  print('Output written to GCS with prefix: {}'.format(gcs_output_uri))

回答1:

It's somewhat unclear from that example, but your call to async_batch_annotate_images takes a requests parameter which is a list of multiple requests. So you can do something like this:

rom google.cloud import vision_v1
from google.cloud.vision_v1 import enums
import six

def generate_request(input_image_uri):
  if isinstance(input_image_uri, six.binary_type):
    input_image_uri = input_image_uri.decode('utf-8')
  if isinstance(output_uri, six.binary_type):
    output_uri = output_uri.decode('utf-8')
  source = {'image_uri': input_image_uri}
  image = {'source': source}
  type_ = enums.Feature.Type.LABEL_DETECTION
  features_element = {'type': type_}
  type_2 = enums.Feature.Type.IMAGE_PROPERTIES
  features_element_2 = {'type': type_2}
  features = [features_element, features_element_2]
  requests_element = {'image': image, 'features': features}

  return requests_element


def sample_async_batch_annotate_images(input_uri, output_uri):
  """Perform async batch image annotation"""

  client = vision_v1.ImageAnnotatorClient()

  requests = [
    generate_request(input_uri.format(filename))
    for filename in os.listdir('my_directory')
  ]

  gcs_destination = {'uri': output_uri}

  # The max number of responses to output in each JSON file
  batch_size = 1
  output_config = {'gcs_destination': gcs_destination, 'batch_size': batch_size}

  operation = client.async_batch_annotate_images(requests, output_config)

  print('Waiting for operation to complete...')
  response = operation.result()

  # The output is written to GCS with the provided output_uri as prefix
  gcs_output_uri = response.output_config.gcs_destination.uri
  print('Output written to GCS with prefix: {}'.format(gcs_output_uri))


sample_async_batch_annotate_images('gs://my_bucket/{}', 'gs://my_bucket/results')

This can annotate up to 2,000 images in a single request. The only downside is that you can only specify a single output_uri as a destination, so you won't be able to use counter to put each result in a separate file, but you can set batch_size = 1 to ensure each response is written separately if this is what you want.

来源：https://stackoverflow.com/questions/58732768/how-to-annotate-multiple-images-from-a-single-call-using-googles-vision-api-py