def initCamera(self):
try:
self.cam = ids.Camera()
self.cam.color_mode = ids.ids_core.COLOR_RGB8 # Get images in RGB format
self.cam.auto_exposure = True
self.cam.continuous_capture = True # Start image capture
return self.OP_OK
except:
return self.OP_ERR
def main():
#
# Sync the clock.
#
sync_time()
#
# get our vehicle
#
#api = local_connect()
#vehicle = api.get_vehicles()[0]
#
# Get the camera
#
dev = ids.Camera()
dev.continuous_capture = False
#
# Configure the camera
#
#dev.pixelclock = dev.pixelclock_range[1]
dev.auto_white_balance = False
dev.color_mode = ids.ids_core.COLOR_RGB8
exposure_us = 100000
dev.auto_exposure = False
dev.exposure = exposure_us * 1e-3
dev.gain = 0
#
# Start capturing images
#
dev.continuous_capture = True
while True:
print 'Capturing Image'
t = datetime.now()
base_name = os.path.join(
BASE_FOLDER,
t.strftime("/home/odroid/test_imgs/%Y%m%d_%H:%M:%S.%f"))
img_name = base_name + '.jpg'
data_name = base_name + '.txt'
dev.next_save(img_name)
data = {
"img_path": img_name,
"Time": t.strftime("%Y%m%d_%H:%M:%S.%f"),
"Exposure": dev.exposure,
#"location": str(vehicle.location)
}
with open(data_name, 'w') as f:
json.dump(data, f)
dev.continuous_capture = False
def __init__(self):
import ids
self._cam = ids.Camera(nummem=1)
self._cam.auto_white_balance = False
self._cam.continuous_capture = False
self._cam.auto_exposure = False
self._cam.exposure = 100
self._cam.color_mode = ids.ids_core.COLOR_BAYER_8
self._cam.gain_boost = False
self._cam.gain_db = 0
def start_cam(ida, expo=0, pixelclock=24):
# turn on cam, note default exposure time and pixelclock
# what should color mode be?
cam = ids.Camera(ida)
cam.color_mode = ids.ids_core.COLOR_RGB8 # preview blue, saved yellow
#cam.color_mode = ids.ids_core.COLOR_BGR8 # preview yellow, saved blue
cam.pixelclock = pixelclock # reduce pxclock to increase expo
if expo > 0:
cam.exposure = expo
elif expo == 0:
cam.exposure = 50
#cam.auto_exposure = True
return cam
def __init__(self, callback=None):
try:
import ids
globals()['ids'] = ids
self.capture_dev = ids.Camera(nummem=10)
#
# Turn off white balance.
#
self.capture_dev.auto_white_balance = False
#
# Image callback.
#
self._callback = callback
except Exception as e:
logging.exception(
'Error while initializing the camera:\n{}'.format(repr(e)))
raise CameraException('Failed initializing the camera')
#
# set the color mode map
#
self._ids_color_map = {
gs.COLOR_RAW: ids.ids_core.COLOR_BAYER_8,
gs.COLOR_RGB: ids.ids_core.COLOR_RGB8
}
#
# Last exposure used for checking if there is a need
# to throw the next image.
#
self._last_settings = None
print('Initializing camera.')
print('%d cycles with %d s between cycles. Each cycle %d frames.' %
(Nc, wait_time_s, Nt))
print('Total number of acquisition %d (%.2f hours).' %
(Nc * Nt, Nc * Nt * wait_time_s / 3600.))
print('Exposure time %d ms.' % exposure)
on = 100
off = 0
p.start(off)
n = 0
while n < Nc:
start = time.clock()
time.sleep(1)
cam = ids.Camera()
cam.color_mode = ids.ids_core.COLOR_MONO_8
cam.exposure = exposure #in ms
ft = ids.ids_core.FILETYPE_PNG
cam.continuous_capture = True
time.sleep(2)
p.ChangeDutyCycle(on)
for t in range(Nt):
filename = name_template % (n, t)
cam.next_save(
filename, filetype=ft,
quality=100) # first two images often corrupted... a bug?
cam.next_save(filename, filetype=ft, quality=100)
print('%d/%d %s' % (n * Nt + t + 1, Nc * Nt, filename))
cam.next_save(filename, filetype=ft, quality=100)
import matplotlib.pyplot as plt
import matplotlib.image as mpimg
import numpy as np
import ids
exposure = 3 #in ms
name_template = 'cycle%03d_t%03d.png'
cam = ids.Camera(handle=1)
#cam.color_mode = ids.ids_core.COLOR_MONO_8
cam.exposure = exposure #in ms
ft = ids.ids_core.FILETYPE_PNG
cam.continuous_capture = True
img, meta = cam.next()
img, meta = cam.next()
img, meta = cam.next()
if img.ndim > 2:
img = img[:, :, 0]
cent = (img.shape[0] / 2, img.shape[1] / 2)
avg = img.mean()
plt.ion()
plt.figure()
plt.imshow(img[cent[0] - 128:cent[0] + 127, cent[1] - 128:cent[1] + 127])
plt.title('Center ROI')
plt.show()
plt.figure()
a = plt.hist(img.ravel(), bins=256, range=(0, 256))
plt.title('Histogram of intensities, mean = %d' % avg)
lines = temp_raw()
temp_output = lines[1].find('t=')
if temp_output != -1:
temp_string = lines[1].strip()[temp_output + 2:]
temp_c = float(temp_string) / 1000.0
return temp_c
n = 0
while n < Nc:
start = time.clock()
p.ChangeDutyCycle(LED_on)
time.sleep(1)
cam1 = ids.Camera()
serial_num1 = cam1.info['serial_num']
if not os.path.exists(serial_num1):
os.makedirs(serial_num1)
cam1.color_mode = ids.ids_core.COLOR_MONO_8
cam1.exposure = exposure #in ms
cam2 = ids.Camera()
serial_num2 = cam2.info['serial_num']
if not os.path.exists(serial_num2):
os.makedirs(serial_num2)
cam2.color_mode = ids.ids_core.COLOR_MONO_8
cam2.exposure = exposure #in ms
cam1.continuous_capture = True
for t in range(Nt):
def setUp(self):
super().setUp()
self.cam = ids.Camera(handle=HANDLE, color=COLOR)
def cam_deffect_detection():
detectionOn = str(raw_input('Deffect detection mode: (on/off)')) or 'on'
if detectionOn == 'on':
print('Detection mode: ON')
elif detectionOn == 'off':
print('Detection mode: OFF')
else:
print('Input not recognized, not gonna detect.')
stopMachineOn = str(raw_input('Stop machine mode: (on/off)')) or 'on'
if stopMachineOn == 'on':
print('Stop mode: ON')
output_port = int(raw_input('RPi low voltage GPIO port: (27)')) or 27
elif stopMachineOn == 'off':
print('Stop mode: OFF')
else:
print('Input not recognized, not gonna stop.')
now = datetime.datetime.now()
cam = ids.Camera()
cam.color_mode = ids.ids_core.COLOR_RGB8 # Get images in RGB format
cam.auto_exposure = True
cam.continuous_capture = True # Start image capture
#for shutters in range(115, 150 ,5):
#for shutters in range(50,60,5):
#shutters /= 10.
#cam.exposure = shutters # Set initial exposure to 5ms
directory = 'teste_domingo_migusta/'
if not os.path.exists(directory):
os.makedirs(directory)
#directory2 = 'imagens/%s' % now.strftime('%Y_%m_%d_%H_%M_%S')
#if not os.path.exists(directory2):
# os.makedirs(directory2)
#print(shutters)
N = int(raw_input('Number of photos: ')) or 15
for i in range(N):
#path1 = directory2 +'/image%s.jpg' % now.strftime('%Y_%m_%d_%H_%M_%S')
#camera.capture(path1)
#print('Saved: ' + path1)
img, meta = cam.next() # Get image as a Numpy array
#pil_img = Image.fromarray(img)
nome_imagem = 'imagem_%s.jpg' % datetime.datetime.now().strftime(
'%Y_%m_%d_%H_%M_%S')
path2 = directory + nome_imagem
#pil_img.save(path2, quality = 100)
misc.imsave(path2, img)
#pil_img.save('teste_domingo_migusta/last_image.jpg', quality = 100)
misc.imsave('teste_domingo_migusta/last_image.jpg', img)
print('Saved: ' + path2)
fabric = {
'_id': i,
'defect': 'None',
'date': datetime.datetime.now(),
'imageUrl': 'imgs/' + nome_imagem,
'deviceID': 'GJjybzAy5V'
}
if detectionOn == 'on':
deffect_lycra = funcao_deteccao_lycra_tracadelas(path2)
deffect_agulha = funcao_detecao_agulhas(path2)
if stopMachineOn == 'on' and (deffect_lycra[0] or deffect_agulha):
if deffect_agulha:
fabric['defect'] = 'Agulha'
print("Defeito agulha!")
if deffect_lycra[0]:
fabric['defect'] = deffect_lycra[1]
print("Defeito lycra!")
GPIO.setmode(GPIO.BCM)
GPIO.setup(output_port, GPIO.OUT, initial=GPIO.LOW)
GPIO.output(output_port, GPIO.LOW)
sleep(1)
GPIO.setup(output_port, GPIO.IN, pull_up_down=GPIO.PUD_DOWN)
#break
#db['fabrics'].insert_one(fabric)
sleep(.8)
import ids
from PyQt4 import QtGui
import cv2
import numpy as np
import matplotlib.pyplot as plt
import ids
from ids.gui import IdsQtApp, IdsQtView
with ids.Camera(nummem=2, color=ids.COLOR_MONO_8) as cam:
#==========================================================================
# Settings
#==========================================================================
cam.color_mode = ids.COLOR_MONO_8 # Get images in RGB format
# cam.exposure = 5 # Set initial exposure to 5ms
cam.auto_exposure = True
cam.auto_exposure_brightness = True
# cam.auto_speed = True
# cam.auto_white_balance = False
# cam.continuous_capture = True # Start image capture
cam.gain = 0
# cam.height = 600
# cam.width = 800
cam.pixelclock = 10
# #==============================================================================
# # TMP
# #==============================================================================
# cam.continuous_capture = True
# img, metadata = cam.next()
# cd = CircleDetector(1)
# cd.process(img)
