def image_analysis_in_stream(subscription_key):
"""ImageAnalysisInStream.
This will analysis an image from a stream and return all available features.
"""
client = ComputerVisionAPI(COMPUTERVISION_LOCATION, CognitiveServicesCredentials(subscription_key))
with open(os.path.join(IMAGES_FOLDER, "house.jpg"), "rb") as image_stream:
image_analysis = client.analyze_image_in_stream(
image_stream,
visual_features=[
VisualFeatureTypes.image_type, # Could use simple str "ImageType"
VisualFeatureTypes.faces, # Could use simple str "Faces"
VisualFeatureTypes.categories, # Could use simple str "Categories"
VisualFeatureTypes.color, # Could use simple str "Color"
VisualFeatureTypes.tags, # Could use simple str "Tags"
VisualFeatureTypes.description # Could use simple str "Description"
]
)
print("This image can be described as: {}\n".format(image_analysis.description.captions[0].text))
print("Tags associated with this image:\nTag\t\tConfidence")
for tag in image_analysis.tags:
print("{}\t\t{}".format(tag.name, tag.confidence))
print("\nThe primary colors of this image are: {}".format(image_analysis.color.dominant_colors))
def recognize_text(subscription_key):
"""RecognizeTextUsingRecognizeAPI.
This will recognize text of the given image using the recognizeText API.
"""
import time
client = ComputerVisionAPI(COMPUTERVISION_LOCATION, CognitiveServicesCredentials(subscription_key))
with open(os.path.join(IMAGES_FOLDER, "make_things_happen.jpg"), "rb") as image_stream:
job = client.recognize_text_in_stream(
image_stream,
mode="Printed",
raw=True
)
operation_id = job.headers['Operation-Location'].split('/')[-1]
image_analysis = client.get_text_operation_result(operation_id)
while image_analysis.status in ['NotStarted', 'Running']:
time.sleep(1)
image_analysis = client.get_text_operation_result(operation_id)
print("Job completion is: {}\n".format(image_analysis.status))
print("Recognized:\n")
lines = image_analysis.recognition_result.lines
print(lines[0].words[0].text) # "make"
print(lines[1].words[0].text) # "things"
print(lines[2].words[0].text) # "happen"
def whatsthat():
try:
now = datetime.datetime.now()
timer = str(now.date()) + str(now.hour) + str(now.minute) + str(
now.second)
name_docu = os.path.join(os.getcwd(), 'tempimages', timer + '.jpeg')
cap = cv2.VideoCapture(0)
r, image = cap.read()
cv2.imwrite(name_docu, image)
cap.release()
# import cv2
# img = cv2.imread('makeharvard.png')
# cv2.imwrite(timer+'.jpeg',img)
source = tinify.from_file(
os.path.join(os.getcwd(), 'tempimages', timer + '.jpeg'))
url = source.url
# Get region and key from environment variables
region = 'westcentralus'
key = '3202352b4d3e49678aa066c513ae0ef2'
# Set credentials
credentials = CognitiveServicesCredentials(key)
# Create client
client = ComputerVisionAPI(region, credentials=credentials)
# url = "http://www.public-domain-photos.com/free-stock-photos-4/travel/san-francisco/golden-gate-bridge-in-san-francisco.jpg"
language = "en"
max_descriptions = str(3)
analysis = client.describe_image(url, max_descriptions, language)
if len(analysis.captions) > 0:
captionn = analysis.captions[0].text
print(captionn)
modular_speech(captionn)
else:
save_speech('unknownError')
except Exception as e:
print(e)
def main():
client = ComputerVisionAPI(
COMPUTER_VISION_API_LOCATION,
CognitiveServicesCredentials(COMPUTER_VISION_API_KEY))
with open('image.jpg', 'rb') as image_stream:
image_analysis = client.analyze_image_in_stream(
image_stream,
visual_features=[
VisualFeatureTypes.tags, VisualFeatureTypes.description
])
print("Image description: {}".format(
image_analysis.description.captions[0].text))
print("Tags:\nTag\t\tConfidence")
for tag in image_analysis.tags:
print("{}\t\t{}".format(tag.name, tag.confidence))
def recognize_printed_text_in_stream(subscription_key):
"""RecognizedPrintedTextUsingOCR_API.
This will do an OCR analysis of the given image.
"""
import time
client = ComputerVisionAPI(COMPUTERVISION_LOCATION, CognitiveServicesCredentials(subscription_key))
with open(os.path.join(IMAGES_FOLDER, "computer_vision_ocr.png"), "rb") as image_stream:
image_analysis = client.recognize_printed_text_in_stream(
image_stream,
language="en"
)
lines = image_analysis.regions[0].lines
print("Recognized:\n")
for line in lines:
line_text = " ".join([word.text for word in line.words])
print(line_text)
def retrieve_text_from_url(imgurl):
client = ComputerVisionAPI(COMPUTERVISION_LOCATION, CognitiveServicesCredentials(COMP_VIS_SUBSCRIPTION_KEY))
txt_analysis2=client.recognize_text(imgurl,raw=True, mode='Printed')
#give Computer Vision some time to process image, could also be a while loop checking status (20s is arbitrary)
time.sleep(20)
#Operation-Location contains url to results, use it to get the processed JSON results
headers = {'Ocp-Apim-Subscription-Key':COMP_VIS_SUBSCRIPTION_KEY}
url = txt_analysis2.response.headers['Operation-Location']
return json.loads(requests.get(url, headers=headers).text)
def whatsthat():
now = datetime.datetime.now()
timer = str(now.date()) + str(now.hour) + str(now.minute) + str(now.second)
with picamera.PiCamera() as camera:
# set camera resolution
camera.resolution = cameraResolution
# print("Starting camera preview...")
camera.capture(timer+ '.jpeg', format='jpeg')
camera.close()
# import cv2
# img = cv2.imread('makeharvard.png')
# cv2.imwrite(timer+'.jpeg',img)
source = tinify.from_file(os.path.join(os.getcwd(), timer + '.jpeg'))
url = source.url
# Get region and key from environment variables
region = 'westcentralus'
key = '43977e2279b849c1bb5c463387b37307'
# Set credentials
credentials = CognitiveServicesCredentials(key)
# Create client
client = ComputerVisionAPI(region, credentials=credentials)
# url = "http://www.public-domain-photos.com/free-stock-photos-4/travel/san-francisco/golden-gate-bridge-in-san-francisco.jpg"
language = "en"
max_descriptions = str(3)
analysis = client.describe_image(url,max_descriptions, language)
captionn = analysis.captions[0].text
modular_speech(captionn)
def retrieve_text_from_img(img):
client = ComputerVisionAPI(COMPUTERVISION_LOCATION, CognitiveServicesCredentials(COMP_VIS_SUBSCRIPTION_KEY))
#raw - returns the direct response alongside the deserialized response
with open(os.path.join(IMAGES_FOLDER, img), "rb") as image_stream:
txt_analysis2=client.recognize_text_in_stream(image_stream,raw=True)
#give Computer Vision some time to process image, could also be a while loop checking status (20s is arbitrary)
time.sleep(20)
#Operation-Location contains url to results, use it to get the processed JSON results
headers = {'Ocp-Apim-Subscription-Key':COMP_VIS_SUBSCRIPTION_KEY}
url = txt_analysis2.response.headers['Operation-Location']
return json.loads(requests.get(url, headers=headers).text)
def readit():
try:
region = 'westcentralus'
key = '3202352b4d3e49678aa066c513ae0ef2'
# Set credentials
credentials = CognitiveServicesCredentials(key)
# Create client
client = ComputerVisionAPI(region, credentials=credentials)
name_docu = 'tempread.jpeg'
cap = cv2.VideoCapture(0)
r, image = cap.read()
cv2.imwrite(name_docu, image)
cap.release()
source = tinify.from_file(os.path.join(os.getcwd(), name_docu))
url = source.url
mode = TextRecognitionMode.handwritten
raw = True
custom_headers = None
numberOfCharsInOperationId = 36
save_speech('waitForText')
# Async SDK call
rawHttpResponse = client.recognize_text(url, mode, custom_headers, raw)
# Get ID from returned headers
operationLocation = rawHttpResponse.headers["Operation-Location"]
idLocation = len(operationLocation) - numberOfCharsInOperationId
operationId = operationLocation[idLocation:]
result = client.get_text_operation_result(operationId)
# SDK call
while result.status in ['NotStarted', 'Running']:
time.sleep(1)
result = client.get_text_operation_result(operationId)
# Get data
main_string = ''
if result.status == TextOperationStatusCodes.succeeded:
for line in result.recognition_result.lines:
main_string = main_string + ' ' + line.text
main_string = re.sub("!|/|;|:|-", "", main_string)
main_string = main_string.replace('|', '')
main_string = main_string.replace('*', '')
modular_speech(main_string)
# sent_token = main_string.split('.')
#
# for sente in sent_token:
# modular_speech(sente)
# if keyboard.is_pressed('d'):
# break
else:
save_speech('unknownError')
except Exception:
pass
def readit():
region = 'westcentralus'
key = '43977e2279b849c1bb5c463387b37307'
# Set credentials
credentials = CognitiveServicesCredentials(key)
# Create client
client = ComputerVisionAPI(region, credentials=credentials)
with picamera.PiCamera() as camera:
# set camera resolution
camera.resolution = cameraResolution
# print("Starting camera preview...")
name_docu = 'tempread.jpeg'
camera.capture(name_docu, format='jpeg')
camera.close()
source = tinify.from_file(os.path.join(os.getcwd(), name_docu + '.jpeg'))
url = source.url
mode = TextRecognitionMode.handwritten
raw = True
custom_headers = None
numberOfCharsInOperationId = 36
save_speech('waitForText')
# Async SDK call
rawHttpResponse = client.recognize_text(url, mode, custom_headers, raw)
# Get ID from returned headers
operationLocation = rawHttpResponse.headers["Operation-Location"]
idLocation = len(operationLocation) - numberOfCharsInOperationId
operationId = operationLocation[idLocation:]
result = client.get_text_operation_result(operationId)
# SDK call
while result.status in ['NotStarted', 'Running']:
time.sleep(1)
result = client.get_text_operation_result(operationId)
# Get data
main_string = ''
if result.status == TextOperationStatusCodes.succeeded:
for line in result.recognition_result.lines:
main_string = main_string + ' ' + line.text
main_string = re.sub("!|/|;|:|-", "", main_string)
main_string = main_string.replace('|','')
main_string = main_string.replace('*', '')
modular_speech(main_string)
# sent_token = main_string.split('.')
# for sente in sent_token:
# modular_speech(sente)
# if GPIO.input(cn1) == 0:
# break
else:
save_speech('unknownError')