def prediction():
if request.method == 'POST':
file = request.files['audiofile']
path = os.path.join(app.config['UPLOAD_FOLDER'], file.filename)
file.save(path)
melSpec = submitAudio(path)
os.remove(path)
results = ci.classify_image(melSpec)
# if melSpec not None:
# prediction = getPrediction( './audio.jpg')
return results
else:
return "hello"
def detect_image(self):
path = "pt.png"
cap = None
if False == self.is_enabled:
cap = cv2.VideoCapture(0)
ret, self.frame = cap.read()
cv2.imwrite(path, self.frame)
if cap != None:
cap.release()
cv2.destroyAllWindows()
c_img = classify_image()
s = self.__detected_object_to_answer(c_img.detect_image(path))
os.remove(path)
if s == "":
return "I am sorry but I don\'t know."
return "This is " + s
print(start_time - time.time(), "secs to start.")
#if current_event["Event"] == "Sleep in bed" and start_time > time.time():
#set_angle(35, servo_channel, pwm)
while (time_left > 0):
ret, frame = cap.read()
#cv2.imshow('frame', frame )
if time.time() - start_time >= time_interval:
frame = frame[:, :, ::-1] #change color from BGR to RGB
image = Image.fromarray(frame)
image = image.resize((width, height)) # resize image to (224, 224)
set_angle(90, servo_channel, pwm)
label_id1, prob1 = classify_image(interpreter1, image)
set_angle(35, servo_channel, pwm)
label_id2, prob2 = classify_image(interpreter2, image)
if labels2[label_id2][2:] == "sleep in bed":
classification_result = "sleep in bed"
else:
classification_result = labels1[label_id1][2:]
start_time = time.time()
time_left -= time_interval
# Return the classification label of the image.
#classification_result = labels1[label_id1][2:]
send_msg = [
"classifiedResult", {
"origin": current_event,
"classified_result": classification_result,
print 'Starting road detector'
if (len(sys.argv) == 3):
if (sys.argv[1] == "train"):
train.train_model(sys.argv[2])
elif (sys.argv[1] == "test"):
image_path = sys.argv[2]
files = os.listdir(image_path)
files = list(filter(lambda x: 'jpg' in x and 'aux' not in x, files))
filenames = list(map((lambda x: re.sub('\.jpg$', '', x)), files))
for file in filenames:
print '-----------------'
print 'classifying image ' + file
path = image_path + file + '.jpg'
result = classify_image.classify_image(path)
cv2.imwrite(file + 'raw.jpg', result)
print 'processing image ' + file
processed_image = extract_paths.extract_paths(result)
cv2.imwrite(file + 'processed.jpg', processed_image)
elif (sys.argv[1] == "post-process"):
image_path = sys.argv[2]
files = os.listdir(image_path)
files = list(filter(lambda x: 'jpg' in x and 'aux' not in x, files))
filenames = list(map((lambda x: re.sub('\.jpg$', '', x)), files))
for file in filenames:
print '-----------------'
print 'processing image ' + file
path = image_path + file + '.jpg'
input_image_from_file = cv2.imread(path, 0)
processed_image = extract_paths.extract_paths(
def hello():
image_url = request.forms.get('image_url')
returnStr = ""
for i in CIMG.classify_image(image_url):
returnStr += i + "<br />"
return returnStr + "<br />" + "<img src=" + image_url + ">"
label_path = "models_and_labels./labels2.txt"
# Read class labels.
labels = load_labels(label_path)
interpreter = Interpreter(model_path)
print("Model Loaded Successfully.")
interpreter.allocate_tensors()
_, height, width, _ = interpreter.get_input_details()[0]['shape']
print("Image Shape (", width, ",", height, ")")
init_time = time.time()
while (True):
ret, frame = cap.read()
cv2.imshow('frame', frame)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
if time.time() - init_time >= 2:
#image = Image.fromarray(cv2.cvtColor(frame,cv2.COLOR_BGR2RGB)) # convert opencv(frame) to PIL(image)
frame = frame[:, :, ::-1]
image = Image.fromarray(frame)
image = image.resize((width, height)) # resize image to (224, 224)
label_id, prob = classify_image(interpreter, image)
init_time = time.time()
# Return the classification label of the image.
classification_label = labels[label_id]
print("Image Label is :", classification_label, ", with Accuracy :",
np.round(prob * 100, 2), "%.")
csv_path = cmdline_args.csv
if cmdline_args.csv == "default":
csv_path = (cmdline_args.folder + "/overview.csv")
with open(csv_path) as csv_file:
csv_reader = csv.reader(csv_file)
line_count = 0
for row in csv_reader:
if line_count == 0:
line_count += 1
else:
split_row = row[0].split(';')
if split_row[1] != 'None' and cmdline_args.skip:
line_count += 1
continue
image_name = f"classified_{line_count}b.jpg"
image_path = (cmdline_args.folder + "\\" + split_row[0] +
".ndpi").replace("/",
"\\").replace("\\\\", "\\")
annotation_path = (cmdline_args.folder + "\\" +
split_row[1]).replace("/", "\\").replace(
"\\\\", "\\")
mask_path = (cmdline_args.folder + "\\" +
split_row[2]).replace("/",
"\\").replace("\\\\", "\\")
classify_image(model, device, image_path, mask_path,
annotation_path,
f"{output_folder}\\{image_name}", patch_size)
line_count += 1
#print("Model Loaded Successfully.")
interpreter.allocate_tensors()
_, height, width, _ = interpreter.get_input_details()[0]['shape']
#print("Image Shape (", width, ",", height, ")")
interpreter2.allocate_tensors()
_, height, width, _ = interpreter2.get_input_details()[0]['shape']
# Load an image to be classified.
image = Image.open(img_folder + "dog.jpg").convert('RGB').resize(
(width, height))
# Classify the image.
time1 = time.time()
label_id, prob = classify_image(interpreter, image)
label_id2, prob2 = classify_image(interpreter2, image)
time2 = time.time()
classification_time = np.round(time2 - time1, 3)
#print("Classificaiton Time =", classification_time, "seconds.")
# Read class labels.
labels = load_labels(label_path)
labels2 = load_labels(label2_path)
# Return the classification label of the image.
classification_label = labels[label_id].split(" ")[1]
classification_label2 = labels2[label_id2].split(" ")[1]
print("Image Label is :", classification_label, ", with Accuracy :",
np.round(prob * 100, 2), "% by interpreter1")
print("Image Label is :", classification_label2, ", with Accuracy :",
np.round(prob2 * 100, 2), "% by interpreter2")