def __init__(self, vs, width=320, height=450, framerate=32):
self.vs = vs
self.root = tki.Tk()
self.framerate = framerate
self.sleepduration = 1.0 / self.framerate
self.frame = None
self.thread = None
self.stopEvent = None
self.root.resizable(width=False, height=False)
self.root.geometry('{}x{}'.format(width, height))
self.panelWidth = width
self.panel = None
self.button = tki.Button(self.root,
text="Ring the Bell!",
command=self.ring)
self.button.pack(side="bottom",
fill="both",
expand="yes",
padx=10,
pady=10)
self.stopVideoLoop = threading.Event()
self.thread = threading.Thread(target=self.videoLoop, args=())
self.thread.start()
self.root.wm_title("Hoosthere")
self.root.wm_protocol("WM_DELETE_WINDOW", self.onClose)
self.recognizer = Recognizer()
def __init__(self):
self.detection_reader = DetectionReader('detections.json')
self.project_file_name = '/home/algernon/andro2'
self.video_file_name = ''
self.db_name = ''
self.data_base = None
self.video_maker = None
self.db_user_name = 'root'
self.db_user_pass = '******'
self.db_host = 'localhost'
self.commands = []
self.output_video_file_name = 'output.mkv'
self.video_reader = None
self.video_writer = None
self.emotion_detection_reader = DetectionReader('emotion_results/er.json')
self.emotion_recognizer = EmotionRecognizer(self.EMOTION_PROB_THRESH)
self.captioner = Captioner('/home/algernon/a-PyTorch-Tutorial-to-Image-Captioning/weights/BEST_checkpoint_coco_5_cap_per_img_5_min_word_freq.pth.tar',
'/home/algernon/a-PyTorch-Tutorial-to-Image-Captioning/weights/WORDMAP_coco_5_cap_per_img_5_min_word_freq.json')
self.segmentator = None
self.clothes_detector = ClothesDetector("yolo/df2cfg/yolov3-df2.cfg", "yolo/weights/yolov3-df2_15000.weights", "yolo/df2cfg/df2.names")
self.face_recognizer = FaceRecognizer()
self.open_project()
self.recognizer = Recognizer(
'/home/algernon/PycharmProjects/AIVlog/mmdetection/configs/pascal_voc/faster_rcnn_r50_fpn_1x_voc0712.py',
'/home/algernon/PycharmProjects/AIVlog/mmdetection/work_dirs/faster_rcnn_r50_fpn_1x_voc0712/epoch_10.pth')
def button_process_click(self):
preprocessor = Preprocessor()
labeled_img = preprocessor.process(self.image)
cv2.imwrite('result_img.png', labeled_img)
recognizer = Recognizer()
disease = recognizer.recognize_disease('result_img.png')
print(disease)
QMessageBox.about(self.mainwindow, "Predicted Disease", disease)
def __init__(self, opts):
#import the recognizer so Gst doesn't clobber our -h
from Recognizer import Recognizer
self.ui = None
self.options = {}
ui_continuous_listen = False
self.continuous_listen = False
self.commander = Command.Commander(command_file,strings_file)
#load the options file
self.load_options()
#merge the opts
for k,v in opts.__dict__.items():
if (not k in self.options) or opts.override:
self.options[k] = v
if self.options['interface'] != None:
if self.options['interface'] == "q":
from QtUI import UI
elif self.options['interface'] == "g":
from GtkUI import UI
elif self.options['interface'] == "gt":
from GtkTrayUI import UI
else:
print "no GUI defined"
sys.exit()
self.ui = UI(args, self.options['continuous'])
self.ui.connect("command", self.process_command)
#can we load the icon resource?
icon = self.load_resource("icon.png")
if icon:
self.ui.set_icon_active_asset(icon)
#can we load the icon_inactive resource?
icon_inactive = self.load_resource("icon_inactive.png")
if icon_inactive:
self.ui.set_icon_inactive_asset(icon_inactive)
if self.options['history']:
self.history = []
#create the recognizer
try:
self.recognizer = Recognizer(lang_file, dic_file, self.options['microphone'] )
except Exception, e:
#no recognizer? bummer
sys.exit()
def Main():
if(Train):
print("Loading CSV...")
machine = MyMachine("./Data/labels.csv")
print("CSV Loaded.")
trainCoulumnName = "id"
valuesColumnName = "breed"
filesType = ".jpg"
TrainingPicturesPathDirectory = "C:\\Users\\hosse\\Desktop\\Dog Breed Project\\Data\\train\\"
TestingPicturesPathDirectoy= "./Data/test/"
trainingImagesNo = 12000 # Should be more than 126 due to __show_25_images() u may get Index out of bound error
callbackLogsPath = "C:\\Users\\hosse\\Desktop\\Dog Breed Project\\Logs\\"
modelsSavingPath = "C:\\Users\\hosse\\Desktop\\Dog Breed Project\\Models\\"
machine.initializer(trainCoulumnName,
valuesColumnName,
filesType,
TrainingPicturesPathDirectory,
TestingPicturesPathDirectoy,
trainingImagesNo,
callbackLogsPath,
modelsSavingPath)
machine.Train()
else:
print("Recognizing...")
imgPath = "7.jpg"
# Predict an Image
modelPath = "C:\\Users\\hosse\\Desktop\\Dog Breed Project\\Models\\ModelNO_10_LOSS_0.74_ACCURACY_0.82_IMG_NO_12000.h5"
CSVLabelsPath = "./Data/labels.csv"
X_ColumnName = "id"
Y_ColumnName = "breed"
imagePath = "C:\\Users\\hosse\\Desktop\\Dog Breed Project\\ImagesToTest\\" + str(imgPath)
trainPicturesPath = "C:\\Users\\hosse\\Desktop\\Dog Breed Project\\Data\\train\\"
filesType = ".jpg"
imgReco = Recognizer(modelPath,CSVLabelsPath,X_ColumnName,Y_ColumnName,trainPicturesPath,filesType)
imgReco.predict(imagePath)
return 0
def __init__(self):
super(MainWindow, self).__init__()
self.recognizer = Recognizer()
self.recognizer.learn()
self.ui = loadUi(os.path.join(THIS_DIR, 'mainwindow.ui'), self)
self.thread = QThread()
try:
self.camera = CameraDevice()
except ValueError:
self.ui.video.setText("Device not found!\n\nIs FFMPEG available?")
else:
self.camera.frame_ready.connect(self.update_video_label)
self.ui.video.setMinimumSize(640 * 2, 480)
self.camera.moveToThread(self.thread)
self.ui.t_max.setValue(0)
self.ui.t_min.setValue(255)
self.ui.s_max.setValue(200)
self.ui.s_min.setValue(3)
self.update_values()
def __init__(self, opts):
#import the recognizer so Gst doesn't clobber our -h
from Recognizer import Recognizer
self.ui = None
#keep track of the opts
self.opts = opts
ui_continuous_listen = False
self.continuous_listen = opts.continuous
self.commands = {}
self.read_commands()
self.recognizer = Recognizer(lang_file, dic_file, opts.microphone)
self.recognizer.connect('finished', self.recognizer_finished)
self.matchTime = 0
self.keywordTimeLimit = opts.keytime #set to 0 to always speak the keyword
self.commandFileTime = 0
#updates language file and commands on start
self.checkCommandFile()
self.commandFileTime = os.path.getmtime(command_file)
#read options
if opts.interface != None:
if opts.interface == "q":
#import the ui from qt
from QtUI import UI
elif opts.interface == "g":
from GtkUI import UI
else:
print "no GUI defined"
sys.exit()
self.ui = UI(args, opts.continuous)
self.ui.connect("command", self.process_command)
#can we load the icon resource?
icon = self.load_resource("icon.png")
if icon:
self.ui.set_icon(icon)
if self.opts.history:
self.history = []
def __init__(self, opts):
# Initialize our ROS node:
rospy.init_node('voice')
# Define our publisher:
self.voice_pub = rospy.Publisher('voice', String, queue_size=1)
#import the recognizer so Gst doesn't clobber our -h
from Recognizer import Recognizer
self.ui = None
#keep track of the opts
self.opts = opts
ui_continuous_listen = False
self.continuous_listen = opts.continuous
self.stringsFileTime = os.path.getmtime(strings_file)
self.commands = {}
self.read_commands()
self.recognizer = Recognizer(lang_file, dic_file, opts.microphone)
self.recognizer.connect('finished', self.recognizer_finished)
self.matchTime = 0
self.keywordTimeLimit = opts.keytime #set to 0 to always speak the keyword
# Update the Language File and Commands?
self.commandFileTime = os.path.getmtime(command_file)
if ((AUTO_UPDATE_CMD_FILE)
or (self.commandFileTime > self.stringsFileTime)):
# Trick the system by making it think we just created the command file:
self.commandFileTime = time.time()
self.checkCommandFile()
#read options
if self.opts.history:
self.history = []
"um": audio.wavread(wav_directory + 'um.wav')[0],
"dois": audio.wavread(wav_directory + 'dois.wav')[0],
}
audio_base[3] = {
"matrix": audio.wavread(wav_directory + 'matrix.wav')[0],
"braveheart": audio.wavread(wav_directory + 'braveheart.wav')[0],
"constantine": audio.wavread(wav_directory + 'constantine.wav')[0],
}
audio_base[4] = {
"dinheiro": audio.wavread(wav_directory + 'dinheiro.wav')[0],
"cartao": audio.wavread(wav_directory + 'cartao.wav')[0],
}
audio_base[5] = {
"finalizar_compra":
audio.wavread(wav_directory + 'finalizar_compra.wav')[0],
"sair": audio.wavread(wav_directory + 'sair.wav')[0],
}
if __name__ == "__main__":
for i in range(6):
recorder = Recorder()
recognizer = Recognizer()
recorder.record(time_to_run=2)
(input_signal1, sampling_rate1, bits1) = audio.wavread('record.wav')
Recognizer.test_audio(audio_base[i], input_signal1, wav_directory)
fs = sampling_rate1
lowcut = 300
highcut = 3400
mo.open("com.lx.jdhg", "com.lx.jdhg/com.ly.lxdr.AppActivity")
time.sleep(4)
mo.click(params['skip2_x'], params['skip2_y'])
mo.click(params['skip_x'], params['skip_y'])
mo.click(params['start_button_x'], params['start_button_y'])
while True:
current = time.time() * 1000
p = os.popen(
'adb shell "dumpsys window | grep mCurrentFocus"') # 启动前检测是否在正确的页面
result = str(p.read())
if not result[:-1].endswith("com.lx.jdhg/com.ly.lxdr.AppActivity}"):
restart_app()
continue
reco = Recognizer(mo.get_screen_shot())
try:
table = reco.find()
path = table.find_path(True)
step_time = time.time() * 1000
index = 0
while index < len(path) - 1:
mo.swipe(path[index][0] + params['main_area_west'],
path[index][1] + params['main_area_north'],
path[index + 1][0] + params['main_area_west'],
path[index + 1][1] + params['main_area_north'])
index += 1
print("滑动耗时 %d ms" % (time.time() * 1000 - step_time, ))
mo.click(params['collect_button_x'], params['collect_button_y'])
time.sleep(0.5)
mo.click(params['next_x'], params['next_y'])
def __init__(self, opts):
#import the recognizer so Gst doesn't clobber our -h
from Recognizer import Recognizer
#set variables
self.ui = None
self.options = {}
ui_continuous_listen = False
self.continuous_listen = False
self.commands = {}
#read the commands
self.load_commands_file()
#load the options file
print("load the options")
self.load_options_file()
#merge the opts
for k, v in opts.__dict__.items():
if (not k in self.options) or opts.override:
self.options[k] = v
# should we be updating?
if self.options['update']:
#make the sentences corpus
self.generate_sentences_corpus()
#run the update stuff
UpdateLanguage.update_language()
if self.options['interface'] != None:
if self.options['interface'] == "q":
from QtUI import UI
elif self.options['interface'] == "g":
from GtkUI import UI
elif self.options['interface'] == "gt":
from GtkTrayUI import UI
else:
print("no GUI defined")
sys.exit()
self.ui = UI(args, self.options['continuous'])
self.ui.connect("command", self.process_command)
#can we load the icon resource?
icon = self.load_resource("icon.png")
if icon:
self.ui.set_icon_active_asset(icon)
#can we load the icon_inactive resource?
icon_inactive = self.load_resource("icon_inactive.png")
if icon_inactive:
self.ui.set_icon_inactive_asset(icon_inactive)
if self.options['history']:
self.history = []
#create the recognizer
try:
self.recognizer = Recognizer(lang_file, dic_file,
self.options['microphone'])
except Exception as e:
print(e)
#no recognizer? bummer
sys.exit()
self.recognizer.connect('finished', self.recognizer_finished)
print("Using Options: ", self.options)
def __init__(self):
'''
variables clarification
definition of variables in this class
'''
self.root = Tk()
self.welcome_label = Label(master=self.root, width=80, height=30)
self.photo = PhotoImage(file='new_welc.png')
self.back_right = Label(master=self.root,
width=50,
height=30,
bg='green')
self.back_left = Label(master=self.root,
width=30,
height=30,
bg='blue')
self.img_label = Label(master=self.welcome_label,
image=self.photo,
height=self.photo.height(),
width=self.photo.width())
self.welc_over_but = Button(master=self.img_label,
text='Get Start',
width=20,
height=1,
command=self.come_in)
self.mfcc_label = Label(master=self.back_right,
width=50,
height=16,
bg='yellow')
self.word_label = Label(master=self.back_left,
width=30,
height=30,
bg='pink')
self.word_list = Listbox(master=self.word_label, width=30, height=10)
self.word_rec_label = Label(master=self.word_label,
height=1,
width=30,
bg='red')
self.word_rec_but = Button(master=self.word_rec_label,
height=1,
width=10,
text='Listen to me',
command=self.listen_word)
self.auto_cover = IntVar()
self.word_args = Label(master=self.word_label,
width=30,
height=6,
bg='blue')
self.word_rec_time = Text(master=self.word_args, width=30, height=1)
self.word_file = Text(master=self.word_args, width=30, height=1)
self.word_listen_status = Label(master=self.word_args,
width=30,
height=4,
bg='white')
self.word_result = Listbox(master=self.word_label, height=13, width=30)
self.mfcc_pic_label = Label(master=self.mfcc_label,
height=16,
width=50)
self.mfcc_pic = None
self.talk_label = Label(master=self.back_right,
width=50,
height=14,
bg='purple')
self.talk_arg = Label(master=self.talk_label, width=50, height=2)
self.lcy = Label(master=self.talk_arg, width=20, height=3)
self.fjw = Label(master=self.talk_arg, width=30, height=3, bg='green')
self.talk_rec_but = Button(master=self.lcy,
width=10,
height=2,
command=self.listen_talk,
text="Let's chat!")
self.talk_auto_cover = IntVar()
self.talk_rec_time = Text(master=self.fjw, width=30, height=1)
self.talk_file = Text(master=self.fjw, width=30, height=1)
self.talk_area = Label(master=self.talk_label, width=50, height=12)
self.talk_history = Text(master=self.talk_area, width=50, height=12)
self.speech_content = Text(master=self.talk_area, width=50, height=2)
self.rc = Recognizer()
self.talker = Talker()
self.root.title('Speech Recognizer')
self.welcome()
from VizGen import *
from Recognizer import Recognizer
if __name__ == "__main__":
alphabet = ['B', 'A', 'D', 'C']
generated_string, characters_dict, original_image = generate_image(
alphabet, 8)
noised_image = noise_image(original_image, sigma=100)
recognizer = Recognizer(original_image, noised_image, alphabet,
characters_dict)
print(original_image.shape[1])
print(noised_image.shape[1])
recognized_string = recognizer.recognize()
recognized_image = concatenate_images(recognized_string, characters_dict)
show_triple_images(original_image, noised_image, recognized_image,
"original image", "noised image", "recognized image")
class Controller():
global language
language = "eng"
global recognizer
recognizer = Recognizer()
global engoperations
engoperations = English_Operations()
global uroperations
uroperations = Urdu_Operations()
def __init__(self):
app = QtWidgets.QApplication(sys.argv)
MainWindow = QtWidgets.QMainWindow()
global ui
ui = Ui_MainWindow()
ui.setupUi(MainWindow)
ui.microbutton.clicked.connect(lambda: self.callistner("button"))
ui.querytextfield.returnPressed.connect(
lambda: self.callistner("textfield"))
ui.actionEnglish.triggered.connect(lambda: self.change_language("eng"))
ui.actionUrdu.triggered.connect(lambda: self.change_language("ur"))
ui.add_label("goti", "Hello! I am Gotti. How can I help you?")
recognizer.talk("Hello! I am Gotti. How can I help you?", "en-uk")
MainWindow.show()
sys.exit(app.exec_())
def change_language(self, lang):
global language
language = lang
def makedecision(self, command):
if language == 'eng':
if 'search' in command:
engoperations.OpenChrome(command)
elif 'launch' in command:
engoperations.LaunchApp(command)
else:
ui.add_label("goti", "Sorry can't understand your command")
recognizer.talk("Sorry can't understand Your command", "en-uk")
else:
if 'تلاش' in command:
uroperations.OpenChrome(command)
elif 'کھولو' in command:
uroperations.LaunchApp(command)
else:
ui.add_label("goti", "معاف کیجئے گا آپ کا حکم سمجھ نہیں آیا")
def callistner(self, who):
if ((who == "button") & (language == "eng")):
ui.changetext("Gouti is Listening..")
text = recognizer.myCommand('en-US')
if text == -1:
text = "your last command couldn\'t be heard.please speak again"
ui.add_label("goti", text)
recognizer.talk(text, "en-uk")
else:
ui.add_label("user", text)
ui.changetext("")
self.makedecision(text)
elif ((who == "textfield") & (language == "eng")):
text = ui.querytextfield.text()
ui.add_label("user", text)
ui.changetext("")
self.makedecision(text)
elif ((who == "button") & (language == "ur")):
ui.changetext("گوٹی سن رہا ہے ...")
text = recognizer.myCommand('ur-PK')
if text == -1:
text = "آپ کا آخری حکم نہیں سنا گیا۔ براہ کرم دوبارہ بولیں"
ui.add_label("goti", text)
else:
ui.add_label("user", text)
ui.changetext("")
self.makedecision(text)
elif ((who == "textfield") & (language == "ur")):
text = ui.querytextfield.text()
ui.add_label("user", text)
ui.changetext("")
self.makedecision(text)
ui.scrolled(ui.scrollArea.verticalScrollBar().maximum())
from exemplary_points import filter_and_getExemplaries
sample_rate = 10
# load data
N_users = 10
for i in range(N_users):
execfile("data/user" + str(i + 1) + ".py")
all_data = [data_user1, data_user2, data_user3, data_user4, data_user5, \
data_user6, data_user7, data_user8, data_user9, data_user10]
# extract exemplary points
filter_and_getExemplaries(all_data, sample_rate)
# make prediction
recognizer = Recognizer(sample_rate)
# The data is from letter "O"
print recognizer.predict_one([[0.00151, 0.01135], [0.01175, 0.02839],
[0.0243, -0.00781], [0.01606, -0.03011],
[-0.01251, -0.02651], [-0.08557, 0.02435],
[-0.05574, 0.02611], [0.00386, 0.01013],
[0.03279, 0.00049], [0.02381, -0.0027],
[-0.0126, -0.00321], [-0.06615, 0.01246],
[-0.06997, 0.00418], [-0.06926, -0.00977],
[-0.0444, -0.01925], [0.00515, -0.02594],
[0.01419, -0.02939], [0.00689, -0.01478],
[0.00534, -0.01004], [-0.00194, -0.01154],
[-0.01755, -0.0048], [-0.0367, -0.0101],
[-0.05585, 0.00492], [-0.03075, 0.01247],
[0.01603, -0.02082], [0.04772, -0.01042],
[0.05997, -0.0237], [0.06643, -0.01772],
from Network import Network
#Parse arguments
ap = argparse.ArgumentParser()
ap.add_argument("-p", "--picamera", type=int, default=1,
help="Use Raspberry Camera")
ap.add_argument("-w", "--width", type=int, default=316,
help="Witdh of the window")
ap.add_argument("-ht", "--height", type=int, default=450,
help="Height of the window")
ap.add_argument("-fr", "--framerate", type=int, default=25,
help="Frame rate of the camera")
opt = vars(ap.parse_args())
#recognizer = Recognizer()
recognizer = Recognizer(modelFile='model.mdl')
network = Network()
#network = Network(endpoint='http://localhost:8000/hoo/')
print('INFO: People: ')
print(recognizer.people)
print("INFO: Launching camera")
vs = VideoStream(usePiCamera=opt["picamera"] > 0).start()
time.sleep(2.0)
view = View(vs, recognizer, network, width=opt["width"], height=opt["height"], framerate=opt["framerate"])
print("INFO: Application started successfully.")
view.root.mainloop()
def __init__(self):
super().__init__()
self.initUI()
self.pathList = []
self.rgzr = Recognizer()
