def test_03_analyze(self):
dog_path = os.path.join(os.path.dirname(__file__), '../../resources/dog.jpg')
giraffe_path = os.path.join(os.path.dirname(__file__),
'../../resources/my-giraffe.jpeg')
with open(dog_path, 'rb') as dog_file, open(giraffe_path, 'rb') as giraffe_files:
analyze_images = self.visual_recognition.analyze(
collection_ids=['684777e5-1f2d-40e3-987f-72d36557ef46'],
features=[AnalyzeEnums.Features.OBJECTS.value],
images_file=[
FileWithMetadata(dog_file),
FileWithMetadata(giraffe_files)
],
image_url=['https://upload.wikimedia.org/wikipedia/commons/thumb/4/47/American_Eskimo_Dog.jpg/1280px-American_Eskimo_Dog.jpg']).get_result()
assert analyze_images is not None
print(json.dumps(analyze_images, indent=2))
def recognize_faces(img_path=None, img_url=None):
"""
Call Watson Face Recognition model, to recognize faces in an input images
:param img_path: local path of input raw image file. Only one of img_path or img_url should be used.
:param img_url: http/https url for input raw image. Only one of img_path or img_url should be used.
:return: List of array, each element is the coordinates of faces, based on which faces can be cropped
"""
try:
assert (bool(img_path and img_url) is False)
assert (bool(img_path or img_url) is True)
except AssertionError:
print(
"One and only one of the img_path or img_url should be specified")
if img_path:
with open(img_path, 'rb') as img_f:
analyze_images = service_v4.analyze(
collection_ids=face_detection_CID,
features=AnalyzeEnums.Features.OBJECTS.value,
images_file=[FileWithMetadata(img_f)]).get_result()
else:
analyze_images = service_v4.analyze(
collection_ids=face_detection_CID,
features=AnalyzeEnums.Features.OBJECTS.value,
image_url=[img_url]).get_result()
# return analyze_images
#print(analyze_images)
return format_face_coords(analyze_images)
def label_predict(self):
with open(self.frame, 'rb') as honda_file:
result = self.visual_recognition.analyze(
collection_ids=[self.model_id],
features=[AnalyzeEnums.Features.OBJECTS.value],
images_file=[FileWithMetadata(honda_file)],
threshold=0.20).get_result()
return result['images'][0]['objects']['collections'][0]['objects']
def analyze_results(path):
try:
with open(path, 'rb') as honda_file:
result = visual_recognition.analyze(
collection_ids=[collect_id],
features=[AnalyzeEnums.Features.OBJECTS.value],
images_file=[FileWithMetadata(honda_file)],
threshold=0.25).get_result()
print("helllo")
print(result)
print(type(result))
return result
except:
print(path)
return None
def get_ine():
apikey = 'i6Ze6gqvd1XrPj624Z-BGbejpQosiVKjnKaW-YVsM1cf'
url = 'https://api.us-south.visual-recognition.watson.cloud.ibm.com/instances/a48f0673-5423-47cf-8295-d5b52a90921c'
collection = '031fbbd1-cc79-4cf0-ac3c-e5ffd0c25cdc'
authenticator = IAMAuthenticator(apikey)
service = VisualRecognitionV4('2021-04-29', authenticator=authenticator)
service.set_service_url(url)
path = '././images/ine_ejemplo.jpg'
with open(path, 'rb') as ine_img:
analyze_images = service.analyze(
collection_ids=[collection],
features=[AnalyzeEnums.Features.OBJECTS.value],
images_file=[FileWithMetadata(ine_img)]).get_result()
print(analyze_images)
response = json.dumps(analyze_images)
return Response(response, mimetype='application/json')
def test_04_training(self):
# create a classifier
my_collection = self.visual_recognition.create_collection(
name='my_test_collection',
description='testing for python').get_result()
collection_id = my_collection.get('collection_id')
assert collection_id is not None
# add images
with open(
os.path.join(
os.path.dirname(__file__),
'../../resources/South_Africa_Luca_Galuzzi_2004.jpeg'),
'rb') as giraffe_info:
add_images_result = self.visual_recognition.add_images(
collection_id,
images_file=[FileWithMetadata(giraffe_info)],
).get_result()
assert add_images_result is not None
image_id = add_images_result.get('images')[0].get('image_id')
assert image_id is not None
# add image training data
training_data = self.visual_recognition.add_image_training_data(
collection_id,
image_id,
objects=[
TrainingDataObject(object='giraffe training data',
location=Location(64, 270, 755, 784))
]).get_result()
assert training_data is not None
# train collection
train_result = self.visual_recognition.train(
collection_id).get_result()
assert train_result is not None
assert train_result.get('training_status') is not None
# delete collection
self.visual_recognition.delete_collection(collection_id)
'2018-03-19',
authenticator=authenticator)
service.set_service_url('https://gateway.watsonplatform.net/visual-recognition/api')
# create a classifier
my_collection = service.create_collection(
name='',
description='testing for python'
).get_result()
collection_id = my_collection.get('collection_id')
# add images
with open(os.path.join(os.path.dirname(__file__), '../resources/South_Africa_Luca_Galuzzi_2004.jpeg'), 'rb') as giraffe_info:
add_images_result = service.add_images(
collection_id,
images_file=[FileWithMetadata(giraffe_info)],
).get_result()
print(json.dumps(add_images_result, indent=2))
image_id = add_images_result.get('images')[0].get('image_id')
# add image training data
training_data = service.add_image_training_data(
collection_id,
image_id,
objects=[
TrainingDataObject(object='giraffe training data',
location=Location(64, 270, 755, 784))
]).get_result()
# train collection
train_result = service.train(collection_id).get_result()
def test_04_objects_and_training(self):
# create a classifier
my_collection = self.visual_recognition.create_collection(
name='my_test_collection',
description='testing for python').get_result()
collection_id = my_collection.get('collection_id')
assert collection_id is not None
# add images
with open(
os.path.join(
os.path.dirname(__file__),
'../../resources/South_Africa_Luca_Galuzzi_2004.jpeg'),
'rb') as giraffe_info:
add_images_result = self.visual_recognition.add_images(
collection_id,
images_file=[FileWithMetadata(giraffe_info)],
).get_result()
assert add_images_result is not None
image_id = add_images_result.get('images')[0].get('image_id')
assert image_id is not None
# add image training data
training_data = self.visual_recognition.add_image_training_data(
collection_id,
image_id,
objects=[
TrainingDataObject(object='giraffe training data',
location=Location(64, 270, 755, 784))
]).get_result()
assert training_data is not None
# list objects metadata
object_metadata_list = self.visual_recognition.list_object_metadata(
collection_id=collection_id).get_result()
assert object_metadata_list is not None
# update object metadata
object_metadata = object_metadata_list.get('objects')[0]
updated_object_metadata = self.visual_recognition.update_object_metadata(
collection_id=collection_id,
object=object_metadata.get('object'),
new_object='updated giraffe training data').get_result()
assert updated_object_metadata is not None
# get object metadata
object_metadata = self.visual_recognition.get_object_metadata(
collection_id=collection_id,
object='updated giraffe training data',
).get_result()
assert object_metadata is not None
assert object_metadata.get('object') == 'updated giraffe training data'
# train collection
train_result = self.visual_recognition.train(
collection_id).get_result()
assert train_result is not None
assert train_result.get('training_status') is not None
# training usage
training_usage = self.visual_recognition.get_training_usage(
start_time='2019-11-01', end_time='2019-11-27').get_result()
assert training_usage is not None
# delete object
self.visual_recognition.delete_object(
collection_id, object='updated giraffe training data')
# delete collection
self.visual_recognition.delete_collection(collection_id)
def test_analyze(self):
endpoint = '/v4/analyze'
url = '{0}{1}'.format(base_url, endpoint)
response = {
"images": [{
"objects": {
"collections": [{
"collection_id":
"collection_id",
"objects": [{
"score": 7.0614014,
"location": {
"top": 1,
"left": 5,
"width": 5,
"height": 2
},
"object": "object"
}, {
"score": 7.0614014,
"location": {
"top": 1,
"left": 5,
"width": 5,
"height": 2
},
"object": "object"
}]
}, {
"collection_id":
"collection_id",
"objects": [{
"score": 7.0614014,
"location": {
"top": 1,
"left": 5,
"width": 5,
"height": 2
},
"object": "object"
}, {
"score": 7.0614014,
"location": {
"top": 1,
"left": 5,
"width": 5,
"height": 2
},
"object": "object"
}]
}]
},
"source": {
"archive_filename": "archive_filename",
"filename": "filename",
"type": "file",
"resolved_url": "resolved_url",
"source_url": "source_url"
},
"errors": {
"code": "invalid_field",
"message":
"The date provided for `version` is not valid. Specify dates in `YYYY-MM-DD` format.",
"more_info":
"https://cloud.ibm.com/apidocs/visual-recognition-v4#versioning",
"target": {
"type": "parameter",
"name": "version"
}
},
"dimensions": {
"width": 6,
"height": 0
}
}, {
"objects": {
"collections": [{
"collection_id":
"collection_id",
"objects": [{
"score": 7.0614014,
"location": {
"top": 1,
"left": 5,
"width": 5,
"height": 2
},
"object": "object"
}, {
"score": 7.0614014,
"location": {
"top": 1,
"left": 5,
"width": 5,
"height": 2
},
"object": "object"
}]
}, {
"collection_id":
"collection_id",
"objects": [{
"score": 7.0614014,
"location": {
"top": 1,
"left": 5,
"width": 5,
"height": 2
},
"object": "object"
}, {
"score": 7.0614014,
"location": {
"top": 1,
"left": 5,
"width": 5,
"height": 2
},
"object": "object"
}]
}]
},
"source": {
"archive_filename": "archive_filename",
"filename": "filename",
"type": "file",
"resolved_url": "resolved_url",
"source_url": "source_url"
},
"errors": {
"code": "invalid_field",
"message":
"The date provided for `version` is not valid. Specify dates in `YYYY-MM-DD` format.",
"more_info":
"https://cloud.ibm.com/apidocs/visual-recognition-v4#versioning",
"target": {
"type": "parameter",
"name": "version"
}
},
"dimensions": {
"width": 6,
"height": 0
}
}],
"trace":
"trace",
"warnings": [{
"code": "invalid_field",
"more_info": "more_info",
"message": "message"
}, {
"code": "invalid_field",
"more_info": "more_info",
"message": "message"
}]
}
responses.add(responses.POST,
url,
body=json.dumps(response),
status=200,
content_type='application/json')
authenticator = IAMAuthenticator('bogusapikey')
service = ibm_watson.VisualRecognitionV4('YYYY-MM-DD',
authenticator=authenticator)
service.set_service_url(base_url)
with open(
os.path.join(os.path.dirname(__file__),
'../../resources/cars.zip'), 'rb') as cars:
detailed_response = service.analyze(
collection_ids=['collection_id1, collection_id2'],
features=[AnalyzeEnums.Features.OBJECTS.value],
images_file=[FileWithMetadata(cars)],
image_url=[
'https://upload.wikimedia.org/wikipedia/commons/thumb/4/47/American_Eskimo_Dog.jpg/1280px-American_Eskimo_Dog.jpg'
],
threshold='0.2')
result = detailed_response.get_result()
assert result is not None
assert len(responses.calls) == 2
import json
from ibm_watson import VisualRecognitionV4
from ibm_watson.visual_recognition_v4 import FileWithMetadata, AnalyzeEnums
from ibm_cloud_sdk_core.authenticators import IAMAuthenticator
apikey = 'YOUR API KEY HERE'
url = 'YOUR URL HERE'
collection = 'YOUR COLLECTION HERE'
authenticator = IAMAuthenticator(apikey)
service = VisualRecognitionV4('2018-03-19', authenticator=authenticator)
service.set_service_url(url)
path = 'Path of your image'
with open(path, 'rb') as mask_img:
analyze_images = service.analyze(
collection_ids=[collection],
features=[AnalyzeEnums.Features.OBJECTS.value],
images_file=[FileWithMetadata(mask_img)]).get_result()
obj = analyze_images['images'][0]['objects']['collections'][0]['objects'][0][
'object']
coords = analyze_images['images'][0]['objects']['collections'][0]['objects'][
0]['location']
from matplotlib import pyplot as plt
cv2.imread(path)
"language": "english",
"route": "p",
"numbers": "8300448682, 9025592443"
}
headers = {'cache-control': "no-cache"}
response = requests.request("GET",
url,
headers=headers,
params=querystring)
print(response.text)
with open(picname + '.jpg', 'rb') as img:
result = visual_recognition.analyze(
collection_ids=["2db6113b-aaae-4924-b424-a1103669251f"],
features=[AnalyzeEnums.Features.OBJECTS.value],
images_file=[FileWithMetadata(img)],
threshold='0.6').get_result()
if (result['images'][0]['objects'] != {}):
with open('Can_Enter.mp3',
'wb') as audio_file: #wb means write bytes
audio_file.write(
text_to_speech.synthesize(
'The Gate will open. You can enter now.',
voice='en-US_AllisonVoice',
accept='audio/mp3').get_result().content)
playsound('Can_Enter.mp3')
else:
with open('Cannot_Enter.mp3',
'wb') as audio_file: #wb means write bytes
def camera():
authenticator = IAMAuthenticator(
apikey="ZQBgX2TPs2CmC3wNLWGEr7X3xDmq42sDsqTZVCsjdke8")
flag = False
obj = VisualRecognitionV4(version='2021-03-25',
authenticator=authenticator)
obj.set_service_url(
service_url=
'https://api.us-south.visual-recognition.watson.cloud.ibm.com/instances/2fa93f7c-36f0-48f9-abb1-15845fbe94e1'
)
while True:
success, img = cap.read()
cv2.imwrite('4.jpg', img)
with open('4.jpg', 'rb') as honda_file, open('4.jpg',
'rb') as dice_file:
result = obj.analyze(
collection_ids=[
"17951ae7-0169-4551-ac19-a3864c7eed65",
'17951ae7-0169-4551-ac19-a3864c7eed65'
],
features=[AnalyzeEnums.Features.OBJECTS.value],
images_file=[
FileWithMetadata(honda_file),
],
threshold=0.15).get_result()
#print(json.dumps(result, indent=2))
img = cv2.imread('4.jpg')
try:
left = result['images'][0]['objects']['collections'][0]['objects'][
0]['location']['left']
top = result['images'][0]['objects']['collections'][0]['objects'][
0]['location']['top']
width = result['images'][0]['objects']['collections'][0][
'objects'][0]['location']['width']
height = result['images'][0]['objects']['collections'][0][
'objects'][0]['location']['height']
img = cv2.imread('4.jpg')
cv2.rectangle(img, (left, top), (left + width, top + height),
(255, 0, 0), 2)
name = result['images'][0]['objects']['collections'][0]['objects'][
0]['object']
with open('data.txt', 'r') as f:
f.write(str(name))
cv2.putText(img, name, (left, top), cv2.FONT_HERSHEY_SIMPLEX, 1,
(255, 0, 0), 2)
cv2.imshow('gasm', img)
k = cv2.waitKey(30) & 0xff
if k == 27: # press 'ESC' to quit
break
cv2.show('gasm', img)
if name == 'gun' and flag == False:
flag = True
SendMail()
except:
cv2.imshow('gasm', img)
# pass
# cv2.show('gasm',img)
cap.release()
# cv2.waitKey(0)
cv2.destroyAllWindows()