reworked image encoding started face_rec

.gitignore

@@ -1,2 +1,4 @@
 
 __pycache__/
+
+DP_UareU_WSDK_223/

Neues Textdokument.txt

@@ -0,0 +1 @@
+java.exe -classpath ".;C:\Program Files\DigitalPersona\U.are.U SDK\Windows\Lib\Java\dpuareu.jar" -Djava.library.path="C:\Program Files\DigitalPersona\U.are.U SDK\Windows\Lib\win32" UareUSampleJava

application/endpoints.py

@@ -28,7 +28,7 @@ class PersonList(Resource):
                     session.add(fp)
 
             personJSON["fingerprints"] = fingerprintsObj
-            personJSON["face"] = ("data:image/png;base64,"+str(lastImage)).encode('utf-8')
+            personJSON["face"] = lastImage
             person = Person(**personJSON)
             session.add(person)
             session.commit()
@@ -118,7 +118,7 @@ class Camera(Resource):
     # provides th function used for the live streams
     class VideoCamera(object):
         """Video stream object"""
-        url = "https://webcam1.lpl.org/axis-cgi/mjpg/video.cgi"
+        url = "http://192.168.178.56:8080/video"
         def __init__(self):
             self.video = cv2.VideoCapture(self.url)
 
@@ -145,7 +145,7 @@ class Camera(Resource):
                 return flask.Response(self.gen(self.VideoCamera()), mimetype='multipart/x-mixed-replace; boundary=frame')
             
             elif type == "still":
-                lastImage1 = base64.b64decode(lastImage.encode())
+                lastImage1 = base64.b64decode(lastImage)
                 return flask.Response(lastImage1,  mimetype='image/png')
 
             return flask.make_response(flask.jsonify({'error': "No idea how you got here"}), 404)
@@ -156,7 +156,7 @@ class Camera(Resource):
     def post(self):
         global lastImage
         try:
-            lastImage = base64.b64encode(self.VideoCamera().get_frame('.png')).decode()
+            lastImage = base64.b64encode(self.VideoCamera().get_frame('.png'))
         except Exception as e:
             print("error: -", e)
             return flask.make_response(flask.jsonify({'error': str(e)}), 404)

application/face_rec.py

@@ -0,0 +1,110 @@
+import face_recognition
+import os
+import cv2
+from db import Session, Person
+import base64
+import numpy as np
+from base64 import decodestring
+import base64
+
+from io import StringIO
+from PIL import Image
+
+KNOWN_FACES_DIR = 'known_faces'
+UNKNOWN_FACES_DIR = 'unknown_faces'
+TOLERANCE = 0.6
+FRAME_THICKNESS = 3
+FONT_THICKNESS = 2
+MODEL = 'hog'  # default: 'hog', other one can be 'cnn' - CUDA accelerated (if available) deep-learning pretrained model
+
+
+# Returns (R, G, B) from name
+def name_to_color(name):
+    # Take 3 first letters, tolower()
+    # lowercased character ord() value rage is 97 to 122, substract 97, multiply by 8
+    color = [(ord(c.lower())-97)*8 for c in name[:3]]
+    return color
+
+def readb64(base64_string):
+    sbuf = StringIO()
+    sbuf.write(base64.b64decode(base64_string))
+    pimg = Image.open(sbuf)
+    return cv2.cvtColor(np.array(pimg), cv2.COLOR_RGB2BGR)
+
+print('Loading known faces...')
+known_faces = []
+known_names = []
+session = Session()
+
+# We oranize known faces as subfolders of KNOWN_FACES_DIR
+# Each subfolder's name becomes our label (name)
+for face, name in session.query(Person.face, Person.lname).all():
+        # Load an image
+        nparr = np.fromstring(base64.b64decode(face), np.uint8)
+        image = cv2.imdecode(nparr, cv2.IMREAD_COLOR)
+        
+        # Get 128-dimension face encoding
+        # Always returns a list of found faces, for this purpose we take first face only (assuming one face per image as you can't be twice on one image)
+        encoding = face_recognition.face_encodings(image)[0]
+
+        # Append encodings and name
+        known_faces.append(encoding)
+        known_names.append(name)
+
+
+print('Processing unknown faces...')
+# Now let's loop over a folder of faces we want to label
+
+image = face_recognition.load_image_file('C:/Users/ofjok/Desktop/1.png')
+
+# This time we first grab face locations - we'll need them to draw boxes
+locations = face_recognition.face_locations(image, model=MODEL)
+
+# Now since we know loctions, we can pass them to face_encodings as second argument
+# Without that it will search for faces once again slowing down whole process
+encodings = face_recognition.face_encodings(image, locations)
+
+# We passed our image through face_locations and face_encodings, so we can modify it
+# First we need to convert it from RGB to BGR as we are going to work with cv2
+image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
+
+# But this time we assume that there might be more faces in an image - we can find faces of dirrerent people
+
+for face_encoding, face_location in zip(encodings, locations):
+
+    # We use compare_faces (but might use face_distance as well)
+    # Returns array of True/False values in order of passed known_faces
+    results = face_recognition.compare_faces(known_faces, face_encoding, TOLERANCE)
+
+    # Since order is being preserved, we check if any face was found then grab index
+    # then label (name) of first matching known face withing a tolerance
+    match = None
+    if True in results:  # If at least one is true, get a name of first of found labels
+        match = known_names[results.index(True)]
+        print(f' - {match} from {results}')
+
+        # Each location contains positions in order: top, right, bottom, left
+        top_left = (face_location[3], face_location[0])
+        bottom_right = (face_location[1], face_location[2])
+
+        # Get color by name using our fancy function
+        color = name_to_color(match)
+
+        # Paint frame
+        cv2.rectangle(image, top_left, bottom_right, color, FRAME_THICKNESS)
+
+        # Now we need smaller, filled grame below for a name
+        # This time we use bottom in both corners - to start from bottom and move 50 pixels down
+        top_left = (face_location[3], face_location[2])
+        bottom_right = (face_location[1], face_location[2] + 22)
+
+        # Paint frame
+        cv2.rectangle(image, top_left, bottom_right, color, cv2.FILLED)
+
+        # Wite a name
+        cv2.putText(image, match, (face_location[3] + 10, face_location[2] + 15), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (200, 200, 200), FONT_THICKNESS)
+
+    # Show image
+    cv2.imshow("1", image)
+    cv2.waitKey(0)
+    cv2.destroyWindow("1")

application/static/main.js

@@ -40,7 +40,7 @@ function loadPersonList(data) {
                 <h6 class="card-subtitle mb-2 text-muted">${item["timestamp"]}</h6>
 
                 <p class="card-text">
-                    <img class="listImg" src="${item["face"]}"></img>
+                    <img class="listImg" src="data:image/png;base64,${item["face"]}"></img>
                     <div class="personalInfo">
                         Gender: ${item["gender"]} <br>
                         YoB: ${item["yob"]} <br>

application/static/render.js

@@ -67,7 +67,7 @@ function renderPersonIdentify(data){
         <div class="card border-light"  onclick="focusPerson(${item["person_id"]})" id="person${item["person_id"]}">
             <div class="card-body">
                 <p class="card-text">
-                    <img class="listImg" src="${item["face"]}"></img>
+                    <img class="listImg" src="data:image/png;base64,${item["face"]}"></img>
                     <div class="personalInfo">
                         Name: ${item["fname"]} ${item["lname"]} <br>
                         Gender: ${item["gender"]} <br>
@@ -84,7 +84,7 @@ function renderPersonIdentify(data){
 
 function renderPersonRight(data){
     string = `
-        <img src="${data["face"]}" id="image-right"> </img>
+        <img src="data:image/png;base64,${data["face"]}" id="image-right"> </img>
 
         <h4 class="heroInfo">
             Gender: ${data["gender"]} <br>

requirements.txt

@@ -2,4 +2,6 @@ flask
 flask_restful
 sqlalchemy
 requests
-flask-restful-swagger-3
+flask-restful-swagger-3
+cmake
+face_recognition

run.py

@@ -3,3 +3,4 @@ from application import app
 app.run(host="localhost", port='5001', debug=True)
 
 
+