update

README.md

@@ -47,11 +47,17 @@ Create the database by issuing the following in the data folder `sqlite3 images.
 Download the [AT&T face database](http://www.cl.cam.ac.uk/research/dtg/attarchive/facedatabase.html) and extract it to `data/images` before the server is started. This is needed to build the initial prediction model.
 
     cd data
-    wget http://www.cl.cam.ac.uk/Research/DTG/attarchive/pub/data/att_faces.tar.Z
+    curl -o att_faces.tar.Z http://www.cl.cam.ac.uk/research/dtg/attarchive/pub/data/att_faces.tar.Z
     tar zxvf att_faces.tar.Z
-    mv att_faces images
+    mv orl_faces images
 
 Copy `haarcascade_frontalface_alt.xml` from `<path to opencv source>/data/haarcascades/` to the data folder.
 
 Run with `python server.py` and browse to http://localhost:8888 when the model has been trained.
 
+
+## notice
+```
+pip install opencv-contrib-python
+```
+to make sure the `face` in cv2

opencv.py

@@ -39,20 +39,20 @@ def crop_faces(img, faces):
 def load_images(path):
   images, labels = [], []
   c = 0
-  print "test " + path
+  print("test " + path)
   for dirname, dirnames, filenames in os.walk(path):
-    print "test"
+    print("test")
     for subdirname in dirnames:
       subjectPath = os.path.join(dirname, subdirname)
       for filename in os.listdir(subjectPath):
         try:
           img = cv2.imread(os.path.join(subjectPath, filename), cv2.IMREAD_GRAYSCALE)
           images.append(np.asarray(img, dtype=np.uint8))
           labels.append(c)
-        except IOError, (errno, strerror):
-          print "IOError({0}): {1}".format(errno, strerror)
-        except:
-          print "Unexpected error:" , sys.exc_info()[0]
+        except IOError as e:
+          print("IOError({0}): {1}".format(e))
+        except Exception as e:
+          print("Unexpected error:" , sys.exc_info()[0])
           raise
       c += 1
     return images, labels
@@ -61,32 +61,33 @@ def load_images_to_db(path):
   for dirname, dirnames, filenames in os.walk(path):
     for subdirname in dirnames:
       subject_path = os.path.join(dirname, subdirname)
-      label = Label.get_or_create(name=subdirname)
-      label.save()
+      label, created = Label.get_or_create(name=subdirname)
+      if created is True:
+        label.save(force_insert=True)
       for filename in os.listdir(subject_path):
         path = os.path.abspath(os.path.join(subject_path, filename))
         logging.info('saving path %s' % path)
-        image = Image.get_or_create(path=path, label=label)
-        image.save()
+        image, created = Image.get_or_create(path=path, label=label)
+        if created is True:
+          image.save(force_insert=True)
 
 def load_images_from_db():
   images, labels = [],[]
   for label in Label.select():
     for image in label.image_set:
       try:
         cv_image = cv2.imread(image.path, cv2.IMREAD_GRAYSCALE)
-        cv_image = cv2.resize(cv_image, (100,100))
+        cv_image = cv2.resize(cv_image, (100, 100))
         images.append(np.asarray(cv_image, dtype=np.uint8))
         labels.append(label.id)
-      except IOError, (errno, strerror):
-       print "IOError({0}): {1}".format(errno, strerror)
+      except IOError as e:
+       print("IOError({0}): {1}".format(e))
   return images, np.asarray(labels)
 
 def train():
   images, labels = load_images_from_db()
-  model = cv2.createFisherFaceRecognizer()
-  #model = cv2.createEigenFaceRecognizer()
-  model.train(images,labels)
+  model = cv2.face.FisherFaceRecognizer_create()
+  model.train(images, labels)
   model.save(MODEL_FILE)
 
 def predict(cv_image):
@@ -95,9 +96,7 @@ def predict(cv_image):
   if len(faces) > 0:
     cropped = to_grayscale(crop_faces(cv_image, faces))
     resized = cv2.resize(cropped, (100,100))
-
-    model = cv2.createFisherFaceRecognizer()
-    #model = cv2.createEigenFaceRecognizer()
+    model = cv2.face.FisherFaceRecognizer_create()
     model.load(MODEL_FILE)
     prediction = model.predict(resized)
     result = {
@@ -160,8 +159,5 @@ def persist(self, cv_image):
 
 if __name__ == "__main__":
   load_images_to_db("data/images")
-  #train()
 
-  print 'done'
-  #predict()
-  #train()
+  print('done')

server.py

@@ -8,20 +8,20 @@
 import uuid
 from PIL import Image
 import time
-import StringIO
+from io import StringIO
 import uuid
 import numpy
 import json
-from tornado.options import define, options
+from tornado.options import options
 import opencv
 
-define("port", default=8888, help="run on the given poort", type=int)
+options.define("port", default=8888, help="run on the given poort", type=int)
 
 class Application(tornado.web.Application):
   def __init__(self):
     handlers = [
-        #(r"/", MainHandler),
-        #(r"/facedetector", FaceDetectHandler),
+        (r"/main", MainHandler),
+        (r"/facedetector", FaceDetectHandler),
         (r"/", SetupHarvestHandler),
         (r"/harvesting", HarvestHandler),
         (r"/predict", PredictHandler),
@@ -114,8 +114,10 @@ def main():
   opencv.train()
   logging.info("Model trained")
   app = Application()
+  options.parse_command_line()
+  print(' -- app is listen on: %s' % options.port)
   app.listen(options.port)
-  tornado.ioloop.IOLoop.instance().start()
+  tornado.ioloop.IOLoop.current().start()
 
 if __name__ == "__main__":
   main()