def process(file_list):
if not classify.CLASSIFIER:
init()
pred_list = classify.run(file_list, classify.CLASSIFIER, classify.LABEL_FILE)
pred_dict = {}
pred_dict["predictions"] = pred_list
return pred_dict
def process(file_list):
if not classify.CLASSIFIER:
init()
pred_list = classify.run(file_list, classify.CLASSIFIER,
classify.LABEL_FILE)
pred_dict = {}
pred_dict['predictions'] = pred_list
return pred_dict
def post(self):
args = self.reqparse.parse_args()
url = args['url']
# Read the image and save it
path = read_image_from_url(url)
print "Image Downloaded to: " + path
# Run the classifier on the image
results = classify.run(path)
# Remove the image after analyzing
remove_image(path)
return results
def profile():
if request.method == "POST":
file = request.files['file']
if file and allowed_file(file.filename):
filename = secure_filename(file.filename)
image_path = os.path.join(app.config['UPLOAD_FOLDER'], filename)
file.save(image_path)
data = run(image_path, cfg)
os.remove(image_path)
return render_template("display.html", data=data)
return "did not work"
return render_template("upload.html")
def upload_file():
if request.method == 'POST':
file = request.files['image']
if file.filename == '':
print('No selected file')
return redirect(request.url)
if file and allowed_file(file.filename):
filename = secure_filename(file.filename)
file.save(os.path.join(app.config['UPLOAD_FOLDER'], filename))
return classify.run(
os.path.join(app.config['UPLOAD_FOLDER'] + "/" + filename))
def test_post():
if request.method == 'POST':
# print(request.get_json())
rev = request.get_json()['Img']
# print(rev)
img_data = base64.b64decode(rev)
with open('001.jpg', 'wb') as f:
f.write(img_data)
# result= classify.run(image)
# print(request.get_data())
res = classify.run(image)
return jsonify({'data': res})
else:
return 'error'
def run(inimage):
#resize image
basewidth = 1080
img = PIL.Image.open(inimage)
wpercent = (basewidth / float(img.size[0]))
hsize = int((float(img.size[1]) * float(wpercent)))
img = img.resize((basewidth, hsize), PIL.Image.ANTIALIAS)
img.save('resized_image.jpg')
#split image
tiles = image_slicer.slice('resized_image.jpg',9)
score=[0,0,0,0,0,0,0,0,0]
#get least quality tile
for tile in tiles:
tile.image.save('tile.jpg')
score[tile.number-1] = float(subprocess.check_output(['./brisque_cpp/brisquequality','-im','tile.jpg']))
index=score.index(max(score))
#get label of image
classlabel=classify.run(inimage).split(' ', 1)[1]
print classlabel
#get classid of label
searchfile = open("labelidmap.txt", "r")
for line in searchfile:
if classlabel in line: classid=int(line.split(':',1)[0])
searchfile.close()
print classid
#perform deepdraw
#tiles[index].image=deepdraw.run(tiles[index].image, classid)
image=deepdraw.run(tiles[index].image,classid)
tiles[index].image=PIL.Image.fromarray(image.astype('uint8'))
#im.save('new.png')
im=image_slicer.join(tiles)
im.save('final.png')
return True
'''
for tile in tiles:
tile.image.show()
'''
''''
#Does something, probably.
if cfg.plot == True:
import plot as p
if cfg.quick == True:
p.run(cfg.name + "quick", quick = True)
else:
p.run(cfg.name)
#Classifies the data into a varying number of categories
if cfg.classify == True:
import classify as c
if cfg.quick == True:
c.run(cfg.name + "quick", cfg.fit, quick = True)
else:
c.run(cfg.name, cfg.fit)
#Creates Histograms for each category identied by learning algorithm
if cfg.onedhistogram == True:
import histogram as h
if cfg.quick == True:
h.run(cfg.name + "quick", int(cfg.groups), str(cfg.fit), quick = True)
else:
h.run(cfg.name, int(cfg.groups), str(cfg.fit))
end = time.time()
print time.asctime(time.localtime()), "Code Ended"
