def run(updateImageSignal, mode, setLabelSignal):
camera.capture('image.jpg')
ogImg = cv2.imread('image.jpg')
colorImg = ogImg.copy()
img = resize(ogImg, 50)
imgContour = img.copy()
img = img[20:200, 0:500]
imgContour = imgContour[20:200, 0:500]
cv2.imwrite('imageCrop.jpg', imgContour)
imgBlur = cv2.GaussianBlur(img, (7, 7), 1)
imgGray = cv2.cvtColor(imgBlur, cv2.COLOR_BGR2GRAY)
threshold1 = 40
threshold2 = 40
imgCanny = cv2.Canny(imgGray, threshold1, threshold2)
kernel = np.ones((5, 5))
imgDil = cv2.dilate(imgCanny, kernel, iterations=1)
form = getContours(imgDil, imgContour, img)
print(form.corners)
imgContour = resize(imgContour, 50)
cv2.imwrite('output.jpg', imgContour)
updateImageSignal.emit()
classifyForm(form, mode, setLabelSignal)
def compare_different_profiles(profile1, profile1_name, profile2_name, profile2, failed_path, not_exist_path,
resolution):
start_time = time.time()
resize.resize(profile2, resolution)
# profile1_loc_faces = get_faces_loc(profile1, ALGORITHM)
# profile1_faces = get_faces(profile1, ALGORITHM, profile1_loc_faces)
# profile2_loc_faces = get_faces_loc(profile2, ALGORITHM)
# profile2_faces = get_faces(profile2, ALGORITHM, profile2_loc_faces)
profile1_faces = dnn.get_faces_from_image(profile1)
profile2_faces = dnn.get_faces_from_image(profile2)
# result,m,n = compare_images(profile1, profile2, ALGORITHM, face_locs1=profile1_loc_faces, face_locs2=profile2_loc_faces,
# faces1=profile1_faces, faces2=profile2_faces)
result,m,n = FN.compare_faces_FN(profile1_faces, profile2_faces)
time_it_took = time.time() - start_time
debug_print("comparing time = " + str(time.time() - start_time))
if result == SAME:
debug_print("Oops. it said 2 different people are ###THE SAME###!")
save_2_images_different_persons(failed_path, profile1, profile2, profile1_name, profile2_name)
save_faces_2_person(failed_path,profile1_faces[m], profile2_faces[n], profile1_name, profile2_name)
return True, time_it_took
elif result == DIFFERENT:
debug_print("Great! it said 2 different people are different!")
elif save:
imsave(not_exist_path + profile2_name, profile2)
return False, time_it_took
def gallery(imageDir, x, y):
print(imageDir)
n = 1000
# filelist = [ f for f in os.listdir(tempPath) if f.endswith(".jpg") ]
# for file in filelist:
# os.remove(tempPath+file)
# for file in os.listdir(imageDir+'/'):
# img1=cv2.imread(imageDir+'/'+file)
# cv2.imwrite(imageDir+str(n)+"pic.jpg",img1[0])
resize.resize(imageDir, x, y)
imgs = []
n_file = 0
filelist = [
f for f in os.listdir(imageDir + "/") if f.endswith(".jpg" or ".png")
]
for file in filelist:
n_file = n_file + 1
imgs.append(Image.open(imageDir + "/" + file))
#img = Image.open("1pic.jpg")
album(imgs).save("album.png")
basewidth = 700
img1 = Image.open("album.png")
wpercent = (basewidth / float(img1.size[0]))
hsize = int((float(img1.size[1]) * float(wpercent)))
img1 = img1.resize((basewidth, hsize), Image.ANTIALIAS)
img1.save("resized_album.png")
def uploadPicContent(self, content):
conn = httplib.HTTPConnection("upload.che168.com",
timeout=timeout_che168)
headers = copy.copy(self.headers)
img = StringIO(content)
smallImg = StringIO()
resize.resize(img, (600, 600), False, smallImg)
content = smallImg.getvalue()
conn.request("POST",
"/UploadImage.ashx?infoid=1000",
content,
headers=headers)
try:
res = conn.getresponse()
resHeaders = res.getheaders()
logger.debug(str(resHeaders))
photoRes = self.decodeBody(resHeaders, res.read())
except Exception as e:
logger.debug(str(e))
ret = None
return ret
#print photoRes#.decode("GB18030")
conn.close()
try:
ret = json.loads(photoRes)
except Exception as e:
#ret = json.loads('[{"msg": "/2014/7/5/5525213967984534785.jpg", "img": "http://www.autoimg.cn/2scimg/2014/7/5/m_5525213967984534785.jpg", "success": 1}]')
logger.debug(str(e))
logger.debug(urllib.quote(photoRes))
ret = None
return ret
def main(args):
in_data_dir = args.in_data_dir
out_data_dir = args.out_data_dir
size = (args.x, args.y)
out_ext = args.out_ext
in_exts = args.in_exts
if not os.path.exists(out_data_dir):
os.mkdir(out_data_dir)
in_images_dir = os.path.join(in_data_dir, "images")
out_images_dir = os.path.join(out_data_dir, "images")
if not os.path.exists(out_images_dir):
os.mkdir(out_images_dir)
names = os.listdir(in_images_dir)
pbar = pb.ProgressBar(widgets=["N=%d|" % len(names), pb.Percentage(), pb.Bar(), pb.ETA()], maxval=len(names))
pbar.start()
for i, name in enumerate(os.listdir(in_images_dir)):
if os.path.splitext(name)[1] in in_exts:
in_path = os.path.join(in_images_dir, name)
out_path = os.path.join(out_images_dir, name)
resize(in_path, out_path, size, out_ext)
pbar.update(i)
pbar.finish()
def main(args):
in_data_dir = args.in_data_dir
out_data_dir = args.out_data_dir
size = (args.x, args.y)
out_ext = args.out_ext
in_exts = args.in_exts
if not os.path.exists(out_data_dir):
os.mkdir(out_data_dir)
in_images_dir = os.path.join(in_data_dir, "images")
out_images_dir = os.path.join(out_data_dir, "images")
if not os.path.exists(out_images_dir):
os.mkdir(out_images_dir)
names = os.listdir(in_images_dir)
pbar = pb.ProgressBar(
widgets=["N=%d|" % len(names),
pb.Percentage(),
pb.Bar(),
pb.ETA()],
maxval=len(names))
pbar.start()
for i, name in enumerate(os.listdir(in_images_dir)):
if os.path.splitext(name)[1] in in_exts:
in_path = os.path.join(in_images_dir, name)
out_path = os.path.join(out_images_dir, name)
resize(in_path, out_path, size, out_ext)
pbar.update(i)
pbar.finish()
def create_los(images, crop, ui=None):
"""
:param images: table of the selected images
:return: the image
Autor : Adrien 2018-01-16
"""
#Dimension of the table
x, y = np.shape(images)
final = np.zeros(((x - 1) * crop, (y - 1) * 2 * crop, 3), np.uint8)
#Borne for y
borne_y = y - 1
##Import and process
# Squares for the odd lines
for i in range(1, x, 2):
for j in range(borne_y):
img = Image.open(images[i][j])
tab = np.array(img)
img.close()
if j == y - 1:
tab = resize.resize(tab, (crop, 2 * crop))
final[(i - 1) * crop:(i + 1) * crop,
j * 2 * crop:(2 * j + 1) * crop] = tab
else:
tab = resize.resize(tab, (2 * crop, 2 * crop))
final[(i - 1) * crop:(i + 1) * crop,
j * 2 * crop:(j + 1) * 2 * crop] = tab
if ui != None:
ui.pb_avance.setValue(
int((100 * ((i * y + j) / (x * y) + 1) / 4 + 50)))
#Triangles for the even lines
for i in range(0, x, 2):
for j in range(y):
img = Image.open(images[i][j])
tab = np.array(img)
img.close()
if i == 0 and j == 0:
beggining = (0, 0)
elif j == 0:
beggining = ((i - 1) * crop, 0)
elif i == 0:
beggining = (0, (2 * j - 1) * crop)
else:
beggining = ((i - 1) * crop, (2 * j - 1) * crop)
final = fill.fill_los(final, tab, crop, beggining, i, x, j, y)
if ui != None:
ui.pb_avance.setValue(
int((100 * ((i) * y + j) / (x * y) + 1) / 4 + 75))
return final
def prediction(filename, slider):
#Original file name for mouse hover on predicted image in predict.html
original_image_path = "../static/uploads/" + filename
#Checking the value of variable 'slider'.
#This is from the html slider in index.html
#to select pre-processing method.
if slider == '1':
type = 'original'
elif slider == '2':
type = 'mirror'
elif slider == '3':
type = 'center'
elif slider == '4':
type = '10crop'
if type == "original":
# Image is read from the uploads folder using the filename from the created url.
image = plt.imread(os.path.join('static/uploads', filename))
resize(filename, False, True)
filename = "distorted" + filename
#Sending image and filename to predict method and getting predictions and image_path in return
predictions, image_path = pred(image, filename)
#Makes a mirror version of the image
elif type == "mirror":
resize(filename, False, True)
filename = "distorted" + filename
image = mirror_image(filename)
filename = "mirror" + filename
# Sending image and filename to predict method and getting predictions and image_path in return
predictions, image_path = pred(image, filename)
#Crops the center of the image. (long sides (height or width) gets cut off, short side (height or width) stays the same).
elif type == "center":
image = resize(filename)
filename = "center_crop" + filename
# Sending image and filename to predict method and getting predictions and image_path in return
predictions, image_path = pred(np.array(image), filename)
elif type == "10crop":
#Sending filename to ten crop method to make 10 versions of the same image
#The probability score gets combined for all 10 versions to make a final/combined probability score
predictions, image_path = ten_crop_pred(filename)
#This will make sure that there are no backslashes in the file path
#html <img src=> Can handle any slash.
#javascript code in <script> tags can not handle paths with backslash.
image_path = image_path.replace('\\', '/')
#return will send user to predict.html (in templates folder) and make the predictions dictionary available in the html code.
return render_template('predict.html', predictions=predictions, image_path=image_path, original_image_path=original_image_path)
def send_email(self):
gmail_user = '******'
gmail_password = self.pw # your gmail password
msg = MIMEMultipart()
msg['Subject'] = 'SYSS Photos'
msg['From'] = gmail_user
msg['To'] = self.receiver
msg.attach(
MIMEText("""
The photos selected are listed below. Thank you for using the system!
If you have any enquiry, please contact Computer Club.
We would appreciate it if you could give me some advice about the User Experience.
Please spare a little time and fill in the questionnaire below.
"""))
msg.attach(
MIMEText(
u'<a href="https://docs.google.com/forms/d/e/1FAIpQLSc98pWBdyFoooQjyEq33VAw3UjJORW6gJIudiN2j5IaP6nwPw/viewform?usp=sf_link">Click Here</a>',
'html'))
msg.attach(
MIMEText(
"""
Kind regards
SYSS Computer Club William and Hardy
""", 'plain', 'utf-8'))
for fl in self.list_of_photos:
resize(fl)
path = fl
while path[-1] != '\\':
path = path[:-1]
att = MIMEImage(open(path + "___temp.jpg", 'rb').read())
name = os.path.basename(fl)
att["Content-Disposition"] = 'attachment; filename=' + name
msg.attach(att)
try:
os.remove(path + "___temp.jpg")
except OSError:
pass
server = smtplib.SMTP_SSL('smtp.gmail.com', 465)
server.ehlo()
try:
server.login(gmail_user, self.pw)
except smtplib.SMTPAuthenticationError:
print("Wrong Password")
return False
server.send_message(msg)
# try:
# smtpObj = smtplib.SMTP('localhost')
# smtpObj.sendmail(sender, receivers, message.as_string())
# except smtplib.SMTPException:
# print("ERROR.")
server.quit()
return True
def _process_jobs(jobs):
for job in jobs:
if not os.path.exists(job.raw_path()):
logging.warning('Could not find image file: %s', job.raw_path())
continue
if os.path.exists(job.resized_path()):
continue
logging.info('Processing %s', job.raw_path())
resize.resize(job)
logging.info('Saved resized image to %s', job.resized_path())
def process(self):
resize.resize(self.path)
temp_path = self.path
while (temp_path[-1]!='\\'):
temp_path = temp_path[:-1]
img = fr.api.load_image_file(temp_path+'___temp.jpg')
location = fr.api.face_locations(img, number_of_times_to_upsample=1, model='hog')
encoding = fr.api.face_encodings(img, known_face_locations=location, num_jitters=1)
for i in range(len(encoding)) :
self.faces.append(face(self.path, location[i], encoding[i]))
def _onProcBtn(self, widget, *data):
inDir = self.inDirEntry.get_text()
outDir = self.outDirEntry.get_text()
resX = self.xResEntry.get_text()
resY = self.yResEntry.get_text()
watermark = self.watermarkEntry.get_text()
self.procBtn.set_sensitive(False)
resize(inDir, outDir, resX, resY, watermark)
self.procBtn.set_sensitive(True)
def post(self):
files = self.__validate_post_args()
if files is None:
return
self.set_status(200)
self.finish()
print('resizing images')
print(files)
resize.resize(files)
def train(self):
print("Image Preprocessing...")
rename()
print("Re-naming images complete!")
resize()
print("Re-sizing images complete!")
print("Training KNN classifier...")
classifier = train("../FakeImageDetection/train",
model_save_path="trained_knn_model.clf",
n_neighbors=1)
print("Training complete!")
def uploadPicContent(self, content, imageConfig):
# {"uploadUrl":"http:\/\/v0.api.upyun.com\/bximg\/",
# "uploadParams":
# {"policy":"",
# "signature":""},
# "vendorName":"upyun",
# "imageSuffix":"#up",
# "fileKey":"file"}
uploadUrl = imageConfig.get('uploadUrl')
params = imageConfig.get('uploadParams')
policy = params.get('policy')
imageSuffix = imageConfig.get('imageSuffix')
signature = params.get('signature')
host = uploadUrl.replace('http://', '')
host = host.replace('/bximg/', '')
logger.debug("Baxing upload image host = " + host)
conn = httplib.HTTPConnection(host, timeout=10)
headers = copy.copy(self.headers)
# jsessionid = self.cookies.get("JSESSIONID", None)
# logger.debug(jsessionid)
img = StringIO(content)
smallImg = StringIO()
resize.resize(img, (800, 600), False, smallImg)
boundaryHeader = "----pluploadboundary" + str(self.millis())
headers[
'Referer'] = 'http://s.baixing.net/swf/uploader/swfupload.swf?preventswfcaching=' + str(
self.millis())
headers['Origin'] = 'http://s.baixing.net'
headers[
"Content-Type"] = "multipart/form-data; boundary=" + boundaryHeader
content = smallImg.getvalue()
boundary = "--" + boundaryHeader
picForm = boundary + '\r\n' + 'Content-Disposition: form-data; name="policy"\r\n\r\n'
picForm += str(policy) + '\r\n'
picForm += boundary + '\r\n' + 'Content-Disposition: form-data; name="signature"\r\n\r\n'
picForm += str(signature) + '\r\n'
picForm += boundary + '\r\n' + 'Content-Disposition: form-data; name="file"; filename="' + str(
self.millis()) + '.jpg"\r\nContent-Type: image/jpeg\r\n\r\n'
picForm += str(content) + '\r\n'
picForm += boundary + "--"
headers['Content-Length'] = len(picForm)
logger.debug('upload image form = ' + picForm)
conn.request("POST", "/bximg/", picForm, headers=headers)
res = conn.getresponse()
result = self.decodeBody(res.getheaders(), res.read())
conn.close()
jsonResult = json.loads(result)
if jsonResult.get('code') == 200:
imgUrl = str(jsonResult.get('url')) + str(imageSuffix)
return imgUrl
return None
def featpyramid(pic,model):
#?????
pyra = {}
padx = math.ceil(model["maxsize"][0][0][0][1])
pady = math.ceil(model["maxsize"][0][0][0][0])
sbin = model["sbin"][0][0][0][0]
interval = model["interval"][0][0][0][0]
sc = 2.0 **(1.0/interval)
imsize = [pic.shape[1],pic.shape[2]]
max_scale = int(1 + np.floor(math.log(min(imsize)/(5.0*sbin))/math.log(sc)))
pyra["feat"] = list(range(int(max_scale + interval)))
pyra["scales"] = np.zeros((max_scale + interval, 1))
pyra["imsize"] = imsize
time = 0
for i in range(interval):
starttime = datetime.datetime.now()
scaled = resize.resize(pic,1.0/sc**i)
endtime = datetime.datetime.now()
tmp = features.features(scaled,sbin/2.0)
time += (endtime - starttime).seconds
size =[tmp.shape[0],tmp.shape[1]+2*pady+2,tmp.shape[2]+2*padx+2]
pyra["feat"][i]=np.zeros(size)
pyra["feat"][i][:,pady+1:size[1]-pady-1,padx+1:size[2]-padx-1] = tmp
pyra["scales"][i] = 2.0/sc**(i)
#starttime = datetime.datetime.now()
tmp = features.features(scaled,sbin)
#endtime = datetime.datetime.now()
#time += (endtime - starttime).seconds
size =[tmp.shape[0],tmp.shape[1]+2*pady+2,tmp.shape[2]+2*padx+2]
pyra["feat"][i+interval]=np.zeros(size)
pyra["feat"][i+interval][:,pady+1:size[1]-pady-1,padx+1:size[2]-padx-1] = tmp
pyra["scales"][i+interval] = 1.0/sc**(i-1)
for j in range(i+interval,max_scale,interval):
starttime = datetime.datetime.now()
scaled = resize.resize(scaled, 0.5)
endtime = datetime.datetime.now()
tmp = features.features(scaled,sbin)
time += (endtime - starttime).seconds
size =[tmp.shape[0],tmp.shape[1]+2*pady+2,tmp.shape[2]+2*padx+2]
pyra["feat"][j+interval]=np.zeros(size)
pyra["feat"][j+interval][:,pady+1:size[1]-pady-1,padx+1:size[2]-padx-1] = tmp
pyra["scales"][j+interval] = 0.5/sc**(i-1)
for i in range(len(pyra["feat"])):
pyra["feat"][i][31,0:pady+1,:]=1
end=pyra["feat"][i].shape
pyra["feat"][i][31,end[1]-padx-1:end[1],:]=1
pyra["feat"][i][31,:,0:padx+1]=1
pyra["feat"][i][31,:,end[2]-pady-1:end[2]]=1
print time
pyra["padx"] = padx
pyra["pady"] = pady
return pyra
def uploadPicContent(self, content):
conn = httplib.HTTPConnection("www.iautos.cn")
headers = copy.copy(self.headers)
img = StringIO(content)
smallImg = StringIO()
resize.resize(img, (800, 600), False, smallImg)
content = smallImg.getvalue()
boundaryHeader = '----WebKitFormBoundary' + str(random.random())
boundary = '--' + boundaryHeader
formStr = ""
formStr += boundary + "\r\n" + 'Content-Disposition: form-data; name="userid"\r\n\r\n' + "114841" + "\r\n"
formStr += boundary + "\r\n" + 'Content-Disposition: form-data; name="photoCount"\r\n\r\n' + "20" + "\r\n"
formStr += boundary + "\r\n" + 'Content-Disposition: form-data; name="uploadurl"\r\n\r\n' + "/shopadmin/uploadimg/simple/" + "\r\n"
formStr += boundary + "\r\n" + 'Content-Disposition: form-data; name="shopid"\r\n\r\n' + "0" + "\r\n"
formStr += boundary + "\r\n" + 'Content-Disposition: form-data; name="btnUrl"\r\n\r\n' + "http://www.iautos.cn/static2013/images/shopadmin/car_photo_hover.jpg" + "\r\n"
formStr += boundary + "\r\n" + 'Content-Disposition: form-data; name="maxSize"\r\n\r\n' + "8388608" + "\r\n"
formStr += boundary + "\r\n" + 'Content-Disposition: form-data; name="is_add_watermark"\r\n\r\n' + "0" + "\r\n"
formStr += boundary + "\r\n" + 'Content-Disposition: form-data; name="brand_id"\r\n\r\n' + "0" + "\r\n"
formStr += boundary + "\r\n" + 'Content-Disposition: form-data; name="DropSelect"\r\n\r\n' + "1" + "\r\n"
formStr += boundary + "\r\n" + 'Content-Disposition: form-data; name="Filename"\r\n\r\n' + "aa.jpg" + "\r\n"
formStr += boundary + "\r\n" + 'Content-Disposition: form-data; name="flashId"\r\n\r\n' + "flash1" + "\r\n"
formStr += boundary + "\r\n" + 'Content-Disposition: form-data; name="cid"\r\n\r\n' + "1" + "\r\n"
formStr += boundary + '\r\n' + 'Content-Disposition: form-data; name="Filedata"; filename="aa.jpg"\r\nContent-Type: application/octet-stream\r\n\r\n'
formStr += str(content) + "\r\n"
formStr += boundary + "--"
headers['Host'] = 'www.iautos.cn'
headers['Content-Length'] = len(formStr)
headers["Content-Type"] = "multipart/form-data; boundary=" + boundaryHeader
logger.debug("picUpload headers=" + str(headers))
conn.request("POST", "/shopadmin/uploadimg/simple/", formStr, headers=headers)
res = conn.getresponse()
resHeaders = res.getheaders()
logger.debug(str(resHeaders))
photoRes = self.decodeBody(resHeaders, res.read())
# print photoRes#.decode("GB18030")
conn.close()
try:
ret = json.loads(photoRes)
except Exception as e:
logger.debug(str(e))
logger.debug(urllib.quote(photoRes))
ret = None
return ret
def main():
print('需要几次检测(第一次检查需要较长时间):')
k = int(input())
#目录设置
model_dir = '/Users/lulu/Desktop/Python/object_detection_on_pi'
image_path = '/Users/lulu/Desktop/Python/object_detection_on_pi'
images = os.listdir(image_path)
#读取标签
index_dict = load_index(model_dir)
#创建graph
print('开始构建TensorFlow图...')
create_graph(model_dir)
sess = tf.Session()
softmax_tensor = sess.graph.get_tensor_by_name('softmax:0')
#开始预测
j = 0
while (j < k):
print('第%d次检测! 按下回车拍摄' % (j + 1))
input()
f = open(image_path + '/keychain.jpg', 'wb')
f.write(catch_pic())
f.close()
result = []
for image in images:
if image[-3:] in ['jpg', 'JPG', 'peg', 'PEG']:
resize(image_path + '/' + image)
image_data = tf.gfile.FastGFile(image_path + '/' + image,
'rb').read()
#输入图像数据，得到softmax概率值（一个shape=(1,1008)的向量）
print('已拍摄')
predictions = sess.run(softmax_tensor,
{'DecodeJpeg/contents:0': image_data})
#(1,1008)->(1008,)
predictions = np.squeeze(predictions)
objects = ''
split = ','
max_predict_num = 2
for i in range(max_predict_num):
if i == max_predict_num - 1:
split = ' 或者 '
objects = objects + split + index_dict[np.where(
predictions == max(predictions))[0][0]] + ' '
predictions[predictions == max(predictions)] = 0
result.append(image + '中可能有 ' + objects[1:-1] + '.')
for yi in result:
print(yi)
j += 1
time.sleep(1.5)
print('退出.')
sess.close()
def run(path, blur, resize, rotate, sharpen, writetext):
if blur != 0:
blurProcess.process(path, blur)
elif not (resize[0] == 0 or resize[1] == 0):
resizeProcess.resize(path, resize[0], resize[1])
elif rotate != 0:
rotateProcess.rotate(path, rotate)
elif sharpen:
sharperProcess.sharpen(path)
elif not writetext is None:
# TODO: add writeText support
click.echo("Write text is not working yet!")
else:
click.echo("No options were selected for the image %s" % path)
def input_value(scn_img):
h1, w1 = scn_img.shape[:2]
scn_img = resize(h1, w1, 640, scn_img)
hr1, wr1 = scn_img.shape[:2]
#==================================================>KPT_SCN
# fast:
# fast = cv2.FastFeatureDetector_create(49)
# kp1 = fast.detect(scn_img,masked_data)
# surf:
detector = cv2.xfeatures2d.SURF_create(400, 5, 5)
kp1, desc = detector.detectAndCompute(scn_img, None)
# for k in kp1:
# x,y=k.pt
# print x,y
#==================================================>DESCRIPTOR_SCN
# freak:
freakExtractor = cv2.xfeatures2d.FREAK_create()
kp1, des1 = freakExtractor.compute(scn_img, kp1)
# savekeyPointsOnImage(scn_img,"input1.jpg" ,kp1,wr1,hr1)
# img4 = cv2.drawKeypoints(scn_img, kp1,None, color=(0,255,255))
# cv2.imwrite('input1.jpg', img4)
return kp1, des1
def cropFacesAndSaveThemIntoASpecificFolder(img, imno, param, name):
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
cv2.imshow("img", img)
faces = face_cascade.detectMultiScale(gray, 1.3, 5)
faceNo = 0
if param == 0:
faceFolder = r'/home/nvidia/Desktop/iSeeYou/' + name
if not os.path.exists(faceFolder):
os.makedirs(faceFolder)
if param == 1:
faceFolder = r'/home/nvidia/Desktop/test/' + str(imno)
if not os.path.exists(faceFolder):
os.makedirs(faceFolder)
for (x, y, w, h) in faces:
cv2.rectangle(img, (x, y), (x + w, y + h), (255, 0, 0), 2)
roi_gray = gray[y:y + h, x:x + w]
roi_color = img[y:y + h, x:x + w]
crop_img = img[y:y + h, x:x + w]
faceNo = faceNo + 1
crop_img = img = resize.resize(crop_img)
#imname = 'imno:' +str(imno)+ 'CroppedFace_no:' + str(faceNo)
imname = str(imno)
status = cv2.imwrite(os.path.join(faceFolder, imname + '.jpeg'),
crop_img)
return
def prepare_images():
# print("Crop circles 2f")
# crop_circles("img/original/2f/", "img/cropped/2f/")
# print("Crop circles 5c")
# crop_circles("img/original/5c/", "img/cropped/5c/")
#
# print("Augment 2f")
# data_augment("img/cropped/2f/", "img/augmented/2f/")
# print("Augment 5c")
# data_augment("img/cropped/5c/", "img/augmented/5c/")
size = (28, 28)
print("Resize 2f")
resize("img/augmented/2f/", "img/resized/2f/", size)
print("Resize 5c")
resize("img/augmented/5c/", "img/resized/5c/", size)
def preprocess_for_csvreader(imagesInDir, labelsInfile, outfile):
outfile = open(outfile, 'w')
# Create a dictionary from the filenames (keys) to the labels (values)
labelsDict = csvToDict(labelsInfile)
for subdir, dirs, files in os.walk(imagesInDir):
for f in files:
if not f.startswith('.'):
# If you want to save the new images uncomment this line and add
# 'writeFile' as the last argument in the call to resize()
# writeFile = open(f+'_new', 'a')
# Convert to grayscale
img = Image.open(os.path.join(subdir, f)).convert('L')
# Resize
img = resize(img, (100, 100), False, False)
# Convert image to array
img = np.asarray(img)
img = img.flatten()
# Append image array to given imagesOutfile CSV file
imagesWriter.writerow(img)
# Append the label to the csv file
label = labelsDict[f]
labelsWriter.writerow(label)
imagesOutfile.close()
labelsOutfile.close()
def input_value(scn_img):
h1, w1 = scn_img.shape[:2]
scn_img = resize(h1, w1, 640, scn_img)
hr1, wr1 = scn_img.shape[:2]
#==================================================>KPT_SCN
# fast:
# fast = cv2.FastFeatureDetector_create(49)
# kp1 = fast.detect(scn_img,masked_data)
# brisk:
detector = cv2.BRISK_create(70, 2, .5)
kp1 = detector.detect(scn_img, None)
# for k in kp1:
# x,y=k.pt
# print x,y
#==================================================>DESCRIPTOR_SCN
# brisk:
kp1, des1 = detector.compute(scn_img, kp1)
savekeyPointsOnImage(scn_img, "input1.jpg", kp1, wr1, hr1)
# img4 = cv2.drawKeypoints(scn_img, kp1,None, color=(0,255,255))
# cv2.imwrite('input1.jpg', img4)
return kp1, des1
def compare_person_to_others_profile(images, person_index, failed_path, num_of_checks, not_exists_path, resolution):
person, person_images, person_name = get_person_from_images(images, person_index)
profile = get_profile(person_images)
profile = resize.resize(profile, resolution)
num_of_people = len(images)
indexes_to_check = get_random_indexes(person_index, num_of_people, num_of_checks)
true_negative = 0
false_positive = 0
checked = len(indexes_to_check)
avg_compare_time = 0
for i in indexes_to_check:
curr_person, curr_person_images, curr_person_name = get_person_from_images(images, i)
check_profile = get_profile(curr_person_images)
compare_result, compare_time = compare_different_profiles(profile, person_name, curr_person_name,
check_profile,
failed_path,
not_exists_path, resolution)
avg_compare_time += compare_time
if compare_result:
false_positive += 1
else:
true_negative += 1
avg_compare_time /= checked
return true_negative, false_positive, avg_compare_time, checked
def test_resize_jpg():
"""Create thumbs from jpg image."""
with TemporaryDirectory() as tmpdir:
files = resize("test/files/Körsbärsträd.jpg", tmpdir, "test")
assert len(files) == 4
for (fname, fpath) in files:
assert isfile(fpath)
with Image.open(fpath) as img:
# Check that all image have the widths they nominally should
assert fname == f"test_{img.width}.jpg"
# Check that images have roughly the same aspect ratio as the
# original
aspect = img.width / img.height
assert aspect > 1.32
assert aspect < 1.33
# Check that there is no exif data, or possibly only the
# orientation
exif = img.getexif()
assert len(exif) == 0 or len(exif) == 1
if len(exif) == 1:
assert 0x0112 in exif
def Resize_File(self):
doThings = False
if (len(self.inputFile.get(ANCHOR)) < 1):
self.label3['text'] = "Select a file"
self.label3['fg'] = 'Red'
elif (len(self.inputFile.get(ANCHOR)) > 1):
self.label3['text'] = " "
self.label3['fg'] = 'Green'
doThings = True
try:
if (doThings == True):
fileName = self.inputFile.get(ANCHOR)
fileSplit = fileName.split(".")
for opt in self.options:
dimensions = opt.get().split("x")
if dimensions[0] == '0':
continue
else:
filePath = os.path.join(self.defaultInput, fileName)
newPath = os.path.join(
self.defaultOutput,
self.Format_Name(fileSplit, dimensions))
scale = (int(dimensions[0]), int(dimensions[1]))
new_img = resize.resize(filePath, scale)
new_img.save(newPath)
self.label3['text'] = 'Done'
self.UpdateListBox()
except AttributeError:
self.label3['text'] = "Select resize options"
self.label3['fg'] = 'Red'
def _download_data(url, timeout=10):
"""Download data from given URL."""
resp = requests.get(url, allow_redirects=False, timeout=timeout)
if resp.status_code != 200:
raise Exception('Error on http: [{}], HTTP status: [{}]'.format(
url, resp.status_code))
return base64.b64encode(resize(resp.content)).decode('utf-8')
def _download_data(url, timeout=30):
"""Download data from given URL."""
resp = requests.get(url, allow_redirects=False, timeout=timeout)
if resp.status_code != 200:
raise Exception('Error on http: [{}], HTTP status: [{}]'.format(
url, resp.status_code))
print('[_download_data] url:[%s] size:[%d]' % (url, len(resp.content)))
return resize(resp.content)
def resize_image(self, src, w, h, file_name):
out = u'{}/{}_{}'.format(self.image_dir, file_name, time.time())
if h is 0 or w is 0:
result = resize.resize_with_specific_ratio(src, out, width=w, height=h)
else:
result = resize.resize(src, out, w, h)
if result:
return out
def main(imgPath):
img = resize(imgPath, (100, 100))
img = img.convert('L')
arr = np.asarray(img)
dithered = dither(arr)
lines = ditherToText(dithered)
output = '\n'.join(lines)
print(output)
with open('output.txt', 'w', encoding='utf-8') as f:
f.write(output)
async def resize_from_file(
file: UploadFile = File(...), scale_pct: float = 50.0):
image = await file.read()
resized, headers = resize(image, scale_pct)
headers.update({
"orig-filename": file.filename,
"content-type": file.content_type,
})
headers.update({k: str(v) for k, v in headers.items()})
return StreamingResponse(resized, media_type="image/png", headers=headers)
def run(updateImageSignal, mode, setLabelSignal):
camera.capture('image.jpg')
ogImg = cv2.imread('image.jpg')
colorImg = ogImg.copy()
img = resize(ogImg, 50)
imgContour = img.copy()
img = img[20:200, 0:500]
imgContour = imgContour[20:200, 0:500]
cv2.imwrite('imageCrop.jpg', imgContour)
form = canny(imgContour)
print(form.corners)
imgContour = resize(imgContour, 50)
cv2.imwrite('output.jpg', imgContour)
updateImageSignal.emit()
classifyForm(form, mode, setLabelSignal)
def handlePhoto(update, context, isGroup):
"""Handle a photo sent in by user"""
message = update.message
user = update.effective_user
usableCaption = message.caption or ''
if len(usableCaption):
usableCaption = usableCaption.strip()
if (isGroup):
if not usableCaption or not len(
usableCaption) >= 8 or not usableCaption[:8] == '/sticker':
return
usableCaption = message.caption[8:].strip()
setName = 'set_' + str(update.effective_chat.id).replace('-', 'm')
setTitle = update.effective_chat.title + ' pack'
setOwnerId = HARCODED_USER_ID_TODO_REPLACE
member = context.bot.getChatMember(update.effective_chat.id,
setOwnerId)
if member.status not in ["member", "creator", "administrator"]:
return
else:
setName = 'set_' + str(user.id).replace('-', 'm')
setTitle = user.first_name + ' pack'
setOwnerId = user.id
if len(usableCaption) > 50:
return update.message.reply_text('caption too long (max 50)')
if message.photo and len(message.photo):
# get full size photo
filePath = getImageFromImageMessage(update, context)
imgPath = resize.resize(filePath)
stickerEmoji = DEFAULT_EMOJI
if usableCaption and emoji.emoji_count(usableCaption[0]) == 1:
stickerEmoji = usableCaption[0]
usableCaption = usableCaption[1:].strip()
#add caption
if usableCaption:
drawOnBottom = len(usableCaption) > 1 and usableCaption[0] == '-'
if (drawOnBottom):
usableCaption = usableCaption[1:].strip()
imgPath = drawText.draw(imgPath, usableCaption, drawOnBottom)
sticker = stickerModule.createSticker(context.bot, setOwnerId, imgPath,
setName, setTitle, stickerEmoji)
if sticker:
update.message.reply_sticker(sticker.file_id)
else:
update.message.reply_text('error creating sticker')
#remove tmp files
os.remove(filePath)
os.remove(filePath.split('.')[0] + '_r.png')
else:
update.message.reply_text('error')
else: min_size = img_w/3 # print min_size gray_img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) # print gray_img.shape cascade_fn='/usr/local/opt/opencv/share/OpenCV/haarcascades/haarcascade_frontalface_alt2.xml' cascade = cv2.CascadeClassifier(cascade_fn) rects = cascade.detectMultiScale(gray_img, scaleFactor=1.01, minNeighbors=4, minSize=(min_size, min_size), flags=cv.CV_HAAR_SCALE_IMAGE) # print len(rects) if len(rects) > 0: face = rects[len(rects)-1] # print img.shape, face r_img, r_face = resize.resize(img, face, resize_w, resize_h, face_ratio) # print r_img.shape, r_face face_x, face_y, face_w, face_h = r_face # cv2.rectangle(r_img, (face_x,face_y), (face_x+face_w,face_y+face_h), (255,0,0),2) # meanshift (segmented_img, labels_img, number_regions) = pms.segment(r_img, spatial_radius=9, range_radius=4.5, min_density=20) # kmeans pixels = reshape(segmented_img, (r_img.shape[0]*r_img.shape[1], r_img.shape[2])) centroids, _ = kmeans(pixels, 4) # four colors will be found qnt, _ = vq(pixels, centroids)
def start_once():
resize()
Key([mod, "shift"], "r", lazy.restart()),
Key([mod, "shift"], "q", lazy.shutdown()),
Key([mod], "r", lazy.spawncmd()),
# Applications
Key([mod], "c", lazy.spawn("firefox")),
Key([mod], "l", lazy.spawn("clementine")),
Key([mod], "Return", lazy.spawn("urxvt")),
# Key([mod], "i", lazy.spawn("ipython qtconsole --profile labwork")),
Key([mod], "i", lazy.spawn("jupyter qtconsole")),
Key([mod, "shift"], "l", lazy.spawn("xscreensaver-command -lock")),
Key([], "Print", lazy.spawn("scrot")),
# Multimedia keys
Key([], "XF86Display", lazy.function(lambda q: resize(toggle=True))),
Key([mod], "s", lazy.function(lambda q: resize())),
Key([mod, "shift"], "s", lazy.function(lambda q: resize(clone=True))),
Key([], "XF86AudioPlay", lazy.spawn("clementine --play-pause")),
Key([], "XF86AudioNext", lazy.spawn("clementine --next")),
Key([], "XF86AudioPrev", lazy.spawn("clementine --prev")),
Key([], "XF86AudioMute", lazy.spawn("/home/sean/.config/qtile/volume.sh mute")),
Key([], "XF86AudioLowerVolume", lazy.spawn("/home/sean/.config/qtile/volume.sh down")),
Key([], "XF86AudioRaiseVolume", lazy.spawn("/home/sean/.config/qtile/volume.sh up"))
# Key([], "XF86MicMute", lazy.spawn("/home/sean/.qtile/volume.sh mic")),
]
# Drag floating layouts.
mouse = [
Drag([mod], "Button1", lazy.window.set_position_floating(),
start=lazy.window.get_position()),
from keras.models import Sequential
from keras.layers.core import Dense, Dropout, Activation
from keras.optimizers import SGD, Adam, RMSprop
from keras.utils import np_utils
from keras.models import model_from_json
from PIL import Image
import scipy.misc as smp
import numpy as np
import resize
image_path = input("Enter image name: ")
#a = image_path.split('/')
#image = a[len(a)-1]
#image_path = "/images/"+image_path
converted = resize.resize(image_path)
model = model_from_json(open('my_model_architecture.json').read())
model.load_weights('my_model_weights.h5')
model.compile(loss='categorical_crossentropy',
optimizer=RMSprop(),
metrics=['accuracy'])
ima = Image.open(converted)
pixels = np.array(ima.getdata())
width, height = ima.size
def start_once():
resize()
call(['xsetroot', '-cursor_name', 'left_ptr'])
