Merge branch 'efe' into yigit

MyCity/app/src/main/java/gq/yigit/mycity/CrashMapFragment.java

@@ -27,7 +27,7 @@ import org.json.JSONException;
 import org.json.JSONObject;
 
 
-public class CrashMapFragment extends Fragment implements WebRequest.responseListener {
+public class CrashMapFragment extends Fragment implements WebRequest.responseListener, GoogleMap.OnCameraIdleListener {
 	// TODO: Rename parameter arguments, choose names that match
 	// the fragment initialization parameters, e.g. ARG_ITEM_NUMBER
 	private static final String ARG_PARAM1 = "param1";
@@ -88,6 +88,9 @@ public class CrashMapFragment extends Fragment implements WebRequest.responseLis
 		mMapView.onResume();
 		activity = this;
 
+		FileActions file_manager = new FileActions();
+		url = file_manager.readFromFile(getContext(),"server.config").trim();
+
 		try {
 			MapsInitializer.initialize(getActivity().getApplicationContext());
 		} catch (Exception e) {
@@ -106,18 +109,11 @@ public class CrashMapFragment extends Fragment implements WebRequest.responseLis
 						googleMap.setMyLocationEnabled(true);
 					}
 				}
-				markerPoints = new ArrayList<LatLng>();
 
 				googleMap.getUiSettings().setCompassEnabled(true);
 				googleMap.getUiSettings().setMyLocationButtonEnabled(true);
 				googleMap.getUiSettings().setRotateGesturesEnabled(true);
-
-				FileActions file_manager = new FileActions();
-				url = file_manager.readFromFile(getContext(),"server.config").trim();
-				String url_crashes = url + "/crashes";
-				WebRequest data_request = new WebRequest(url_crashes,true,new HashMap<String, String>(),0);
-				data_request.addListener(activity);
-				data_request.execute();
+				googleMap.setOnCameraIdleListener(activity);
 
 				CameraPosition cameraPosition = new CameraPosition.Builder().target(new LatLng(39.925533,32.866287)).zoom(13).build();
 				googleMap.animateCamera(CameraUpdateFactory.newCameraPosition(cameraPosition));
@@ -127,8 +123,21 @@ public class CrashMapFragment extends Fragment implements WebRequest.responseLis
 		return rootView;
 	}
 
+	public void onCameraIdle() {
+		LatLng center = googleMap.getCameraPosition().target;
+		String url_crashes = url + "/crashes";
+		HashMap<String,String> params = new HashMap<>();
+		params.put("lat", String.valueOf(center.latitude));
+		params.put("lng", String.valueOf(center.longitude));
+		WebRequest data_request = new WebRequest(url_crashes,false,params,0);
+		data_request.addListener(this);
+		data_request.execute();
+	}
+
 	public void receivedResponse(boolean success,String response, int reqid){
 		if(success){
+			markerPoints = new ArrayList<LatLng>();
+			googleMap.clear();
 			try{
 				JSONObject crashes = new JSONObject(response);
 				Iterator<String> iter = crashes.keys();

client_side/app.py

@@ -7,7 +7,7 @@ app = Flask(__name__)
 app.config['SECRET_KEY'] = 'yigit007'
 socketio = SocketIO(app)
 
-src_path , file_list = "../server_side/api/modules/databases/",["denunciations","complaints"]
+src_path , file_list = "../server_side/api/modules/databases/",["denunciations","complaints","crashes"]
 changed = {}
 for file in file_list:
 	changed[file] = os.stat(os.path.join(src_path,file+".json"))
@@ -42,7 +42,7 @@ def handle_my_custom_namespace_event():
 
 
 @socketio.on("check",namespace="/complaints_socket")
-def denunciation_handle(msg):
+def compaint_handle(msg):
 
 	change,data = file_check("complaints")
 	if change:
@@ -59,6 +59,22 @@ def handle_my_custom_namespace_event():
 
 	emit("new", json.dumps(json_data), namespace="/complaints_socket")
 
+@socketio.on('check', namespace='/crashes_socket')
+def crash_handle(msg):
+	change, data = file_check("crashes")
+	if change:
+		emit("new", data, namespace='/crashes_socket')
+
+@socketio.on('connect', namespace='/crashes_socket')
+def handle_crash():
+	print("[INFO]: Received socket connection!")
+
+	src = os.path.join(src_path, 'crashes.json')
+	with open(src, 'r') as f:
+		json_data = json.loads(f.read())
+
+	emit("new", json.dumps(json_data), namespace="/crashes_socket")
+
 @app.route('/gui/<path:path>')
 def send_img(path):
 	return send_from_directory('interface/UserData', path)

client_side/interface/UserData/crashes.html

@@ -0,0 +1,311 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <title>Crashes</title>
+    <meta name="viewport" content="initial-scale=1.0">
+    <meta charset="utf-8">
+    <script src="https://ajax.googleapis.com/ajax/libs/jquery/3.3.1/jquery.min.js"></script>
+    <script src="socket.io/node_modules/socket.io-client/dist/socket.io.js"></script>
+    <style>
+        #map {
+            height: 100%;
+        }
+
+        html, body {
+            height: 100%;
+            margin: 0;
+            padding: 0;
+        }
+    </style>
+
+    <style>
+
+        #img:hover{
+            color: #424242;
+            -webkit-transition: all .3s ease-in;
+            -moz-transition: all .3s ease-in;
+            -ms-transition: all .3s ease-in;
+            -o-transition: all .3s ease-in;
+            transition: all .3s ease-in;
+            opacity: 1;
+            transform: scale(1.15);
+            -ms-transform: scale(1.15); /* IE 9 */
+            -webkit-transform: scale(1.15); /* Safari and Chrome */
+            max-width: 400px;
+            max-height: 400px;
+        }
+
+        #img-user{
+            max-width: 150px;
+            max-height: 150px;
+        }
+
+        .title{
+            text-align: left;
+        }
+
+        td{
+            padding: 20px 10px;
+        }
+
+        .popup{
+            display: none; /* Hidden by default */
+            position: fixed; /* Stay in place */
+            left: 0;
+            top: 0;
+            width: 100%; /* Full width */
+            height: 100%; /* Full height */
+            overflow: auto; /* Enable scroll if needed */
+            background-color: rgb(0,0,0); /* Fallback color */
+            background-color: rgba(0,0,0,0.4); /* Black w/ opacity */
+            padding-top: 60px;
+        }
+
+        #table,#table-user{
+            background: white;
+            padding: 40px 60px;
+        }
+
+        #content, #user-content{
+            display: inline-block;
+            margin-top: 125px;
+            background: white;
+        }
+        #submit{
+            top: 50%;
+            transform: translateY(-50%);
+        }
+
+    </style>
+
+</head>
+<body>
+<audio id="audio" src="http://www.soundjay.com/button/beep-07.wav" autoplay="false"></audio>
+<div id="map"></div>
+
+
+
+<!-- ANANA SOKIYIM JAVASCRIPT
+        Sen nası gereksiz bir dilsin amk bütün bi gecemi yedin
+         ben senenin ebeni sikiyim inşallah seni yazan programcının götüne av tüfeği girer
+         AYNISI SOCKET.IO IÇIN DE GEÇERLİ ALLAHIM KAFAYI YİCEM ADAMLAR HERALDE "ACABA NE KADAR ÖZÜRLÜ BI FRAMERWORK YAZARIZ DUYE BAŞLAMIŞLAR IŞE"-->
+
+
+<div id = "popup" align = "center" class = "popup">
+
+    <div id = "content">
+
+        <form action="." id="submission_form">
+
+            <table border="0" cellpadding="10" cellspacing="0" id = "table" class = "modal">
+                <tr>
+                    <td rowspan="2" colspan="7"><img src="cityscape.png" alt="img"id = "img"></td>
+                    <td colspan="2" class="title"><strong>Priority: </strong></td>
+                    <td colspan="1" id = "priority">7</td>
+                </tr>
+                <tr>
+                    <td colspan="2" class="title"><strong>Type: </strong></td>
+                    <td colspan="1" id = "type"> Ambulance</td>
+                </tr>
+                <tr>
+                    <td colspan="4" class="title"><strong>Submitter: </strong></td>
+                    <td colspan="6">
+                        <a href="#" id = "user" class = "panel" onclick=
+                                "document.getElementById('popup').style.display='none';
+                                document.getElementById('user-popup').style.display='block';
+                                setUser(this.getAttribute('userid'))"
+
+                        >Efe Aydın </a>
+                    </td>
+                </tr>
+                <tr>
+                    <td colspan="4" class="title"><strong>Message: </strong></td>
+                    <td colspan="6" id = "message">Test ambulance</td>
+                </tr>
+                <tr>
+                    <td colspan="4" class="title"><strong>Tags: </strong></td>
+                    <td colspan="6" id = "tags">Test tags</td>
+                </tr>
+                <tr>
+                    <td colspan="4" class="title"><strong>Message: </strong></td>
+                    <td colspan="6" id = "input">  <textarea name="message" rows="5" cols="30" id="response">Please enter response.</textarea>
+                    </td>
+                </tr>
+                <tr>
+                    <td colspan="10">
+                        <input type="submit" id = "submit">
+                    </td>
+                </tr>
+            </table>
+        </form>
+
+    </div>
+
+
+</div>
+
+<div id = "user-popup" align = "center" class="popup">
+
+    <div id = "user-content">
+        <table border="0" cellpadding="10" cellspacing="0" id = "table-user" class = "modal">
+
+            <tr>
+                <td rowspan="3" colspan="7"><img src="cityscape.png" alt="img"id = "img-user"></td>
+                <td colspan="2" class="title"><strong>Name: </strong></td>
+                <td colspan="1" id = "name">Efe Aydın</td>
+            </tr>
+            <tr>
+                <td colspan="2" class="title"><strong>TC: </strong></td>
+                <td colspan="1" id = "tc">111111111111</td>
+            </tr>
+            <tr>
+                <td colspan="2" class="title"><strong>Trust: </strong></td>
+                <td colspan="1" id = "trust">7</td>
+            </tr>
+
+            <tr>
+                <td colspan="4" class="title"><strong>Phone: </strong></td>
+                <td colspan="6" id="phone">
+                    0 555 555 55 55
+                </td>
+            </tr>
+            <tr>
+                <td colspan="4" class="title"><strong>E-mail: </strong></td>
+                <td colspan="6" id = "email">efeaydin@citizenview.com</td>
+            </tr>
+            <tr>
+                <td colspan="4" class="title"><strong>Health: </strong></td>
+                <td colspan="6" id = "health">Test tags</td>
+            </tr>
+        </table>
+        <button onclick=
+                        "document.getElementById('popup').style.display='block';
+                            document.getElementById('user-popup').style.display='none';"
+                style="margin-bottom: 5px;">OK</button>
+    </div>
+
+</div>
+
+<script>
+    function setUser(id) {
+        if(!(id in users)) {
+            $.get("https://" + document.domain + ":5000/users/" + id, function (dataStr, status) {
+                data_user = $.parseJSON(dataStr)
+                users[id]=data_user
+                document.getElementById("name").innerHTML = users[id]["realname"];
+                document.getElementById("tc").innerHTML = users[id]["TC"];
+                document.getElementById("trust").innerHTML = users[id]["trustability"];
+                document.getElementById("phone").innerHTML = users[id]["tel"];
+                document.getElementById("email").innerHTML = users[id]["email"];
+                document.getElementById("img-user").setAttribute("src","https://" + document.domain + ":5000"+users[id]["avatar"])
+            });
+        }else {
+            document.getElementById("name").innerHTML = users[id]["realname"];
+            document.getElementById("tc").innerHTML = users[id]["TC"];
+            document.getElementById("trust").innerHTML = users[id]["trustability"];
+            document.getElementById("phone").innerHTML = users[id]["tel"];
+            document.getElementById("email").innerHTML = users[id]["email"];
+            document.getElementById("img-user").setAttribute("src","https://" + document.domain + ":5000" + users[id]["avatar"])
+        }
+    }
+</script>
+
+<script>
+    var modal = document.getElementById('popup');
+
+    window.onclick = function(event) {
+        if (event.target == modal) {
+            modal.style.display = "none";
+        }
+    }
+</script>
+
+<script>
+    var map;
+    function initMap() {
+        map = new google.maps.Map(document.getElementById('map'), {
+            center: {lat: 39.92, lng: 32.85},
+            zoom: 13
+        });
+    }
+</script>
+
+<script>
+    var sound = document.getElementById("audio");
+    var markers = [];
+
+    var socket = io.connect('http://' + document.domain + ':' + location.port + "/crashes_socket");
+    console.log('http://' + document.domain + ":4000/crashes_socket")
+    console.log(socket);
+
+    var markers = [];
+    var denunciation_count = markers.length;
+    var users = {};
+    setInterval(function () {
+        socket.emit("check",{})
+    },1000);
+    var LatLng = null;
+    socket.on("new",function (dataStr) {
+        for (var i = 0; i < markers.length; i++) {
+            markers.pop().setMap(null);
+        }
+        console.log("Update received!")
+        var data = $.parseJSON(dataStr);
+        for(var key in data) {
+            console.log("aaa aaa")
+            data[key].forEach(function (element,index) {
+                console.log("aaa bbb")
+                if(element["location"]["longitude"] === undefined || element["location"]["latitude"] === undefined){
+                    return;
+                }
+                console.log("aaa ccc")
+                LatLng = {lat: parseFloat(element["location"]["latitude"]), lng: parseFloat(element["location"]["longitude"])};
+                console.log("aaa ddd")
+                var marker = new google.maps.Marker({
+                    position: {lat: parseFloat(element["location"]["latitude"]), lng: parseFloat(element["location"]["longitude"])},
+                    map: map,
+                    label: element["priority"].toString(),
+                    title: element["message"]
+                });
+                console.log("aaa eee")
+                marker.addListener('click', function () {
+                    document.getElementById('popup').style.display = 'block';
+                    document.getElementById("priority").innerHTML = element["response"]["priority"];
+                    if(element["response"]["status"]){
+                        document.getElementById("response").innerHTML = element["response"]["message"];
+                    }
+                    document.getElementById("submission_form").setAttribute("index",index);
+                    document.getElementById("submission_form").setAttribute("id",key);
+                    document.getElementById("type").innerHTML = "Road Damage";
+                    document.getElementById("message").innerHTML = element["content"];
+
+                    if (!(key in users)) {
+                        $.get("https://" + document.domain + ":5000/users/" + key, function (dataStr, status) {
+                            data_user = $.parseJSON(dataStr);
+                            users[key] = data_user;
+                            document.getElementById("user").innerHTML = users[key]["realname"];
+                        });
+                    } else {
+                        document.getElementById("user").innerHTML = users[key]["realname"];
+                    }
+                    document.getElementById("user").setAttribute("userid", key);
+                    document.getElementById("img").setAttribute("src", "data:image/png;base64," + element["img"])
+                });
+
+            });
+        }
+        map.setCenter(LatLng);
+
+        if (denunciation_count < markers.length) {
+            sound.play();
+        }
+        denunciation_count = markers.length;
+    });
+
+</script>
+
+<script src="https://maps.googleapis.com/maps/api/js?key=AIzaSyBuOC03IHPA_6TPnfk18b0SAgD1uge4-dk&callback=initMap" async defer></script>
+
+</body>
+</html>

server_side/api/app.py

@@ -4,6 +4,7 @@ from flask_cors import CORS, cross_origin
 
 from modules import user_info, voting_system, rating_system, denunciation, navigation, bus_stops, announcements, complaint, car_crash
 from modules import utility
+from modules import user_set
 
 app = Flask(__name__)
 api = Api(app)

server_side/api/modules/ai_services.py

@@ -0,0 +1,133 @@
+from flask import Flask, request, Response
+from flask_restful import Resource, Api
+
+import os
+from object_detection.utils import label_map_util
+from object_detection.utils import visualization_utils as vis_util
+from object_detection.utils import ops as utils_ops
+from PIL import Image
+import base64
+import io
+import json
+
+import tensorflow as tf
+import sys
+import numpy as np
+
+switches = {"coco":1, "damage":1}
+COCO_MODEL_NAME = "rfcn_resnet101_coco_2018_01_28"
+PATH_TO_FROZEN_COCO_GRAPH = 'modules/'+COCO_MODEL_NAME + '/frozen_inference_graph.pb'
+PATH_TO_FROZEN_DAMAGE_GRAPH =  'modules/trainedModels/ssd_mobilenet_RoadDamageDetector.pb'
+
+
+if sys.platform == "win32":
+    detection_graph_coco = tf.Graph()
+    detection_graph_damage = tf.Graph()
+    with detection_graph_coco.as_default():
+        od_graph_def = tf.GraphDef()
+        with tf.gfile.GFile(PATH_TO_FROZEN_COCO_GRAPH, 'rb') as fid:
+            serialized_graph = fid.read()
+            od_graph_def.ParseFromString(serialized_graph)
+            tf.import_graph_def(od_graph_def, name='')
+    with detection_graph_damage.as_default():
+        od_graph_def = tf.GraphDef()
+        with tf.gfile.GFile(PATH_TO_FROZEN_DAMAGE_GRAPH, 'rb') as fid:
+            serialized_graph = fid.read()
+            od_graph_def.ParseFromString(serialized_graph)
+            tf.import_graph_def(od_graph_def, name='')
+
+    def load_image_into_numpy_array(image):
+        (im_width, im_height) = image.size
+        return np.array(image.getdata()).reshape(
+            (im_height, im_width, 3)).astype(np.uint8)
+
+def run_inference_for_single_image(image, graph,type):
+    global switches
+    global sess_coco
+    global sess_damage
+    with graph.as_default():
+        if(switches[type]):
+            if type == "coco":
+                sess_coco = tf.Session()
+            elif type == "damage":
+                sess_damage = tf.Session()
+            switches[type] = 0
+        if type == "coco":
+            ops = tf.get_default_graph().get_operations()
+            all_tensor_names = {output.name for op in ops for output in op.outputs}
+            tensor_dict = {}
+            for key in [
+                'num_detections', 'detection_boxes', 'detection_scores',
+                'detection_classes', 'detection_masks'
+            ]:
+                tensor_name = key + ':0'
+                if tensor_name in all_tensor_names:
+                    tensor_dict[key] = tf.get_default_graph().get_tensor_by_name(
+                        tensor_name)
+            if 'detection_masks' in tensor_dict:
+                # The following processing is only for single image
+                detection_boxes = tf.squeeze(tensor_dict['detection_boxes'], [0])
+                detection_masks = tf.squeeze(tensor_dict['detection_masks'], [0])
+                # Reframe is required to translate mask from box coordinates to image coordinates and fit the image size.
+                real_num_detection = tf.cast(tensor_dict['num_detections'][0], tf.int32)
+                detection_boxes = tf.slice(detection_boxes, [0, 0], [real_num_detection, -1])
+                detection_masks = tf.slice(detection_masks, [0, 0, 0], [real_num_detection, -1, -1])
+                detection_masks_reframed = utils_ops.reframe_box_masks_to_image_masks(
+                    detection_masks, detection_boxes, image.shape[1], image.shape[2])
+                detection_masks_reframed = tf.cast(
+                    tf.greater(detection_masks_reframed, 0.5), tf.uint8)
+                # Follow the convention by adding back the batch dimension
+                tensor_dict['detection_masks'] = tf.expand_dims(
+                    detection_masks_reframed, 0)
+            image_tensor = tf.get_default_graph().get_tensor_by_name('image_tensor:0')
+            # Run inference
+            output_dict = sess_coco.run(tensor_dict,
+                                   feed_dict={image_tensor: image})
+            # all outputs are float32 numpy arrays, so convert types as appropriate
+            output_dict['num_detections'] = int(output_dict['num_detections'][0])
+            output_dict['detection_classes'] = output_dict[
+                'detection_classes'][0].astype(np.int64)
+            output_dict['detection_boxes'] = output_dict['detection_boxes'][0]
+            output_dict['detection_scores'] = output_dict['detection_scores'][0]
+            if 'detection_masks' in output_dict:
+                output_dict['detection_masks'] = output_dict['detection_masks'][0]
+        elif type=="damage":
+            image_tensor = graph.get_tensor_by_name('image_tensor:0')
+            # Each box represents a part of the image where a particular object was detected.
+            detection_boxes = graph.get_tensor_by_name('detection_boxes:0')
+            # Each score represent how level of confidence for each of the objects.
+            # Score is shown on the result image, together with the class label.
+            detection_scores = graph.get_tensor_by_name('detection_scores:0')
+            detection_classes = graph.get_tensor_by_name('detection_classes:0')
+            num_detections = graph.get_tensor_by_name('num_detections:0')
+            # Actual detection.
+            (boxes, scores, classes, num) = sess_damage.run(
+                [detection_boxes, detection_scores, detection_classes, num_detections],
+                feed_dict={image_tensor: image})
+
+            output_dict = {'detection_classes': np.squeeze(classes).astype(np.int32), 'detection_scores': np.squeeze(scores)}
+
+    return output_dict
+
+
+class Process(Resource):
+    def post(self):
+        base64_img = request.form['img']
+        image = Image.open(io.BytesIO(base64.b64decode(base64_img)))
+        type = request.form["type"]
+        image_np = load_image_into_numpy_array(image)
+        image_np_expanded = np.expand_dims(image_np, axis=0)
+        if type == "coco":
+            output_dict = run_inference_for_single_image(image_np_expanded, detection_graph_coco,type)
+        elif type == "damage":
+            output_dict = run_inference_for_single_image(image_np_expanded, detection_graph_damage,type)
+
+
+        return json.dumps(output_dict,cls=NumpyEncoder)
+
+
+class NumpyEncoder(json.JSONEncoder):
+    def default(self, obj):
+        if isinstance(obj, np.ndarray):
+            return obj.tolist()
+        return json.JSONEncoder.default(self, obj)

server_side/api/modules/car_crash.py

@@ -13,10 +13,12 @@ import os
 import io
 import itertools
 import pickle
-
-
-from object_detection.utils import visualization_utils as vis_util
+import copy
+from urllib.parse import urlencode
+from urllib.request import Request, urlopen
+import ssl
 from object_detection.utils import label_map_util
+import face_recognition
 
 VEHICLE_CLASSES = [3, 6, 8]
 MIN_AREA_RATIO = 0.9
@@ -25,10 +27,12 @@ import numpy as np
 MIN_SCORE_THRESH = 0.6
 
 if sys.platform == "win32":
-    import tensorflow as tf
-
     sys.path.insert(0, r'C:\Users\Tednokent01\Downloads\MyCity\traffic_analyzer')
 
+PATH_TO_LABELS = os.path.join('object_detection/data', 'mscoco_label_map.pbtxt')
+
+
+category_index = label_map_util.create_category_index_from_labelmap(PATH_TO_LABELS, use_display_name=True)
 
 app = Flask(__name__)
 api = Api(app)
@@ -40,53 +44,39 @@ with open(db_path, 'r') as f:
 users_path = os.path.join(app.root_path, 'databases', 'users.json')
 with open(users_path, 'r') as f:
     users = json.load(f)
-PATH_TO_LABELS = '../../traffic_analyzer/object_detection/data/mscoco_label_map.pbtxt'
-PATH_TO_CKPT = '../../traffic_analyzer/rfcn_resnet101_coco_2018_01_28/frozen_inference_graph.pb'
-category_index = label_map_util.create_category_index_from_labelmap(PATH_TO_LABELS, use_display_name=True)
-if sys.platform == "win32":
-    # PATH_TO_LABELS = '../../traffic_analyzer/object_detection/data/kitti_label_map.pbtxt'
-    # PATH_TO_CKPT = 'modules/faster_rcnn_resnet101_kitti_2018_01_28/frozen_inference_graph.pb'
-
-    detection_graph = tf.Graph()
-    with detection_graph.as_default():
-        od_graph_def = tf.GraphDef()
-        with tf.gfile.GFile(PATH_TO_CKPT, 'rb') as fid:
-            serialized_graph = fid.read()
-            od_graph_def.ParseFromString(serialized_graph)
-            tf.import_graph_def(od_graph_def, name='')
-
-    def load_image_into_numpy_array(image):
-        (im_width, im_height) = image.size
-        return np.array(image.getdata()).reshape(
+
+def load_image_into_numpy_array(image):
+    (im_width, im_height) = image.size
+    return np.array(image.getdata()).reshape(
             (im_height, im_width, 3)).astype(np.uint8)
 
+context = ssl._create_unverified_context()
+
+def find_name(image):
+        known_faces = []
+        known_face_names = []
+        for v in users.values():
+            known_faces.append(np.array(v['face_encoding']))
+            known_face_names.append(v['realname'])
+
+        face_encoding = face_recognition.face_encodings(image)[0]
+        results = face_recognition.compare_faces(known_faces, face_encoding)
+        name = "Unknown"
+        face_distances = face_recognition.face_distance(known_faces, face_encoding)
+        best_match_index = np.argmin(face_distances)
+        if results[best_match_index]:
+            name = known_face_names[best_match_index]
+
+        return name
+
 def process_img(img_base64):
-    if sys.platform == "win32":
-        img = Image.open(io.BytesIO(base64.b64decode(img_base64)))
-        with detection_graph.as_default():
-            with tf.Session(graph=detection_graph) as sess:
-                # Definite input and output Tensors for detection_graph
-                image_tensor = detection_graph.get_tensor_by_name('image_tensor:0')
-                # Each box represents a part of the image where a particular object was detected.
-                detection_boxes = detection_graph.get_tensor_by_name('detection_boxes:0')
-                # Each score represent how level of confidence for each of the objects.
-                # Score is shown on the result image, together with the class label.
-                detection_scores = detection_graph.get_tensor_by_name('detection_scores:0')
-                detection_classes = detection_graph.get_tensor_by_name('detection_classes:0')
-                num_detections = detection_graph.get_tensor_by_name('num_detections:0')
-                # the array based representation of the image will be used later in order to prepare the
-                # result image with boxes and labels on it.
-                image_np = load_image_into_numpy_array(img)
-                # Expand dimensions since the model expects images to have shape: [1, None, None, 3]
-                image_np_expanded = np.expand_dims(image_np, axis=0)
-                # Actual detection.
-                (boxes, scores, classes, num) = sess.run(
-                    [detection_boxes, detection_scores, detection_classes, num_detections],
-                    feed_dict={image_tensor: image_np_expanded})
-                # Visualization of the results of a detection.
-
-
-        output_dict = {'detection_classes': np.squeeze(classes).astype(np.int32), 'detection_scores': np.squeeze(scores), 'detection_boxes': np.squeeze(boxes)}
+    if 1:
+        url = 'https://10.10.26.161:5000/ai' # Set destination URL here
+        post_fields = {'img': img_base64,"type":"coco"}     # Set POST fields here
+        request = Request(url, urlencode(post_fields).encode())
+        data = urlopen(request, context=context).read().decode("ascii")
+        output_dict = json.loads(json.loads(data))
+        image_np = load_image_into_numpy_array(Image.open(io.BytesIO(base64.b64decode(img_base64))))
     else:
         with open('image_1_data.pkl', 'rb') as f:
             output_dict = pickle.load(f)
@@ -121,10 +111,7 @@ def process_img(img_base64):
         output_dict_processed["detection_scores"].append(output_dict["detection_scores"][box.index])
         output_dict_processed["detection_boxes"].append(output_dict["detection_boxes"][box.index])
 
-    output_dict_processed["detection_classes"] = np.array( output_dict_processed["detection_classes"])
-    output_dict_processed["detection_scores"] = np.array(output_dict_processed["detection_scores"])
-    output_dict_processed["detection_boxes"] = np.array(output_dict_processed["detection_boxes"])
-
+    people = {}
     for index, i in enumerate(output_dict['detection_classes']):
         score = output_dict['detection_scores'][index]
         if score > MIN_SCORE_THRESH:
@@ -146,22 +133,36 @@ def process_img(img_base64):
                 box = output_dict['detection_boxes'][index]
                 (left, right, top, bottom) = (box[1] * im_width, box[3] * im_width,
                                               box[0] * im_height, box[2] * im_height)
+                person = image_np[top:bottom,left:right]
+
                 if right-left > bottom-top:
-                    injured_people += 1
+                    face_locs = face_recognition.face_locations(person)
+                    name = find_name(person)
+                    people[index] = [0, face_locs, name]
+                else:
+                    face_locs = face_recognition.face_locations(person)
+                    name = find_name(person)
+                    people[index] = [1, face_locs, name]
+
+
     _, buffer = cv2.imencode('.jpg', image_np)
-    image_process = image_np[:]
-    vis_util.visualize_boxes_and_labels_on_image_array(
-            image_process,
-            output_dict_processed["detection_boxes"],
-            output_dict_processed["detection_classes"],
-            output_dict_processed["detection_scores"],
-            category_index,
-            min_score_thresh=MIN_SCORE_THRESH,
-            use_normalized_coordinates=True,
-            line_thickness=8)
-    cv2.imshow("a",image_process)
-    cv2.waitKey(0)
-    return base64.b64encode(buffer).decode('ascii'), cars_involved, injured_people
+#    image_process = image_np[:]
+#    vis_util.visualize_boxes_and_labels_on_image_array(
+#            image_process,
+#            output_dict_processed["detection_boxes"],
+#            output_dict_processed["detection_classes"],
+#            output_dict_processed["detection_scores"],
+#            category_index,
+#            min_score_thresh=MIN_SCORE_THRESH,
+#            use_normalized_coordinates=True,
+#            line_thickness=8)
+#    cv2.imshow("a",image_process)
+#    cv2.waitKey(0)
+
+    for i in range(len(output_dict_processed["detection_classes"])):
+        output_dict_processed["detection_classes"][i] = category_index[output_dict_processed["detection_classes"][i]]["name"]
+
+    return base64.b64encode(buffer).decode('ascii'), cars_involved, injured_people,output_dict_processed,people
 
 
 class Crash(Resource):
@@ -171,7 +172,8 @@ class Crash(Resource):
         id = request.form['id']
         lat, long = request.form['lat'], request.form['long']
 
-        image, car_count, injured = process_img(base64_img)
+        image, car_count, injured,out,people = process_img(base64_img)
+        print(people)
         priority = car_count + injured
         if priority > 10:
             priority = 10
@@ -187,7 +189,8 @@ class Crash(Resource):
             'location': {
                 'latitude': lat,
                 'longitude': long
-            }
+            },
+            "output_dict": out
         }
         if id in crashes:
             crashes[id].append(crash)
@@ -196,16 +199,28 @@ class Crash(Resource):
 
         with open(db_path, 'w') as f:
             json.dump(crashes, f, indent=4)
-
-        cv2.imshow("a",load_image_into_numpy_array(Image.open(io.BytesIO(base64.b64decode(image)))))
-        cv2.waitKey(0)
-
-
         return crash
 
 class Crashes(Resource):
-    def get(self):
-        return crashes
+    def post(self):
+        process_dict = copy.deepcopy(crashes)
+        return_dict = {}
+        for id in process_dict:
+            for i in range(len(process_dict[id])):
+                del process_dict[id][i]["img"]
+
+        for id in process_dict:
+            for i in range(len(process_dict[id])):
+                location = process_dict[id][i]['location']
+                lat, lng = float(request.form['lat']), float(request.form['lng'])
+                if abs(float(location['latitude']) - lat) < 0.3 and abs(float(location['longitude']) - lng) < 0.3:
+                    if id in return_dict:
+                        return_dict[id].append(process_dict[id][i])
+                    else:
+                        return_dict[id] = [process_dict[id][i]]
+
+        return return_dict
+
 
 class Box:
     def __init__(self,coords, type,index):

server_side/api/modules/complaint.py

@@ -8,6 +8,10 @@ from PIL import Image
 import sys
 import datetime
 import cv2
+import ssl
+from urllib.parse import urlencode
+from urllib.request import Request, urlopen
+
 
 if sys.platform == "win32":
 	import tensorflow as tf
@@ -15,13 +19,15 @@ if sys.platform == "win32":
 	import pickle
 
 	sys.path.insert(0, r'C:\Users\Tednokent01\Downloads\MyCity\traffic_analyzer')
-	from utils import label_map_util
+	from object_detection.utils import label_map_util
 
-	from utils import visualization_utils as vis_util
+	from object_detection.utils import visualization_utils as vis_util
 
 app = Flask(__name__)
 api = Api(app)
 
+context = ssl._create_unverified_context()
+
 score_dict = {
 	1: 1,
 	2: 1,
@@ -38,21 +44,11 @@ with open("modules/databases/complaints.json","r") as f:
 
 if sys.platform == "win32":
 	# Path to frozen detection graph. This is the actual model that is used for the object detection.
-	PATH_TO_CKPT =  'modules/trainedModels/ssd_mobilenet_RoadDamageDetector.pb'
 
 	# List of the strings that is used to add correct label for each box.
 	PATH_TO_LABELS = 'modules/trainedModels/crack_label_map.pbtxt'
 
 	NUM_CLASSES = 8
-
-	detection_graph = tf.Graph()
-	with detection_graph.as_default():
-		od_graph_def = tf.GraphDef()
-		with tf.gfile.GFile(PATH_TO_CKPT, 'rb') as fid:
-			serialized_graph = fid.read()
-			od_graph_def.ParseFromString(serialized_graph)
-			tf.import_graph_def(od_graph_def, name='')
-
 	label_map = label_map_util.load_labelmap(PATH_TO_LABELS)
 	categories = label_map_util.convert_label_map_to_categories(label_map, max_num_classes=NUM_CLASSES, use_display_name=True)
 	category_index = label_map_util.create_category_index(categories)
@@ -65,39 +61,26 @@ def load_image_into_numpy_array(image):
 def process_img(img_base64):
 
 	if sys.platform == "win32":
-		img = Image.open(io.BytesIO(base64.b64decode(img_base64)))
-		with detection_graph.as_default():
-			with tf.Session(graph=detection_graph) as sess:
-				# Definite input and output Tensors for detection_graph
-				image_tensor = detection_graph.get_tensor_by_name('image_tensor:0')
-				# Each box represents a part of the image where a particular object was detected.
-				detection_boxes = detection_graph.get_tensor_by_name('detection_boxes:0')
-				# Each score represent how level of confidence for each of the objects.
-				# Score is shown on the result image, together with the class label.
-				detection_scores = detection_graph.get_tensor_by_name('detection_scores:0')
-				detection_classes = detection_graph.get_tensor_by_name('detection_classes:0')
-				num_detections = detection_graph.get_tensor_by_name('num_detections:0')
-				# the array based representation of the image will be used later in order to prepare the
-				# result image with boxes and labels on it.
-				image_np = load_image_into_numpy_array(img)
-				# Expand dimensions since the model expects images to have shape: [1, None, None, 3]
-				image_np_expanded = np.expand_dims(image_np, axis=0)
-				# Actual detection.
-				(boxes, scores, classes, num) = sess.run(
-						[detection_boxes, detection_scores, detection_classes, num_detections],
-						feed_dict={image_tensor: image_np_expanded})
-				# Visualization of the results of a detection.
-				vis_util.visualize_boxes_and_labels_on_image_array(
-						image_np,
-						np.squeeze(boxes),
-						np.squeeze(classes).astype(np.int32),
-						np.squeeze(scores),
-						category_index,
-						min_score_thresh=0.3,
-						use_normalized_coordinates=True,
-						line_thickness=8)
-
-		output_dict = {'detection_classes': np.squeeze(classes).astype(np.int32), 'detection_scores': np.squeeze(scores)}
+
+		url = 'https://127.0.0.1:5000/ai' # Set destination URL here
+		post_fields = {'img': img_base64,"type":"damage"}     # Set POST fields here
+
+		request = Request(url, urlencode(post_fields).encode())
+		img = load_image_into_numpy_array(Image.open(io.BytesIO(base64.b64decode(img_base64))))
+
+		output_dict = json.loads(json.loads(urlopen(request, context=context).read()))
+		print(output_dict)
+		vis_util.visualize_boxes_and_labels_on_image_array(
+				img,
+				np.array(output_dict['detection_boxes']),
+				output_dict['detection_classes'],
+				output_dict['detection_scores'],
+				category_index,
+				instance_masks=output_dict.get('detection_masks'),
+				use_normalized_coordinates=True,
+				line_thickness=8,
+				min_score_thresh=0.3
+		)
 		defects = []
 		for index, i in enumerate(output_dict['detection_classes']):
 			score = output_dict['detection_scores'][index]
@@ -111,8 +94,11 @@ def process_img(img_base64):
 		if priority > 10:
 			priority = 10
 
-		_, buffer = cv2.imencode('.jpg', image_np)
-		return base64.b64encode(buffer).decode('ascii'),priority,defects
+		buffered = io.BytesIO()
+		img = Image.fromarray(img, 'RGB')
+		img.save(buffered, format="JPEG")
+		img_str = base64.b64encode(buffered.getvalue())
+		return img_str.decode("ascii"),priority,defects
 
 	return img_base64, 7,["unprocessed"]
 
@@ -155,4 +141,3 @@ class ComplaintsUpdate(Resource):
 		with open('modules/databases/complaints.json', 'w') as complaints_file:
 			json.dump(complaints, complaints_file, indent=4)
 		return
-

server_side/api/modules/user_set.py

@@ -0,0 +1,36 @@
+import os
+import sys
+import json
+import base64
+import face_recognition
+
+from modules import utils
+
+with open('modules/databases/users.json') as f:
+	users = json.load(f)
+
+for file in os.listdir("images"):
+	if file.endswith(".png") or file.endswith(".jpg"):
+		uid = file.split('.')[0]
+		if len(uid) == 32 and utils.find_by_id(users.values(), uid):
+			full_path = os.path.join("images", file)
+			image = face_recognition.load_image_file(full_path)
+			with open(full_path, 'rb') as f:
+				base64_image = base64.b64encode(f.read())
+
+			if sys.platform == "win32":
+				face_locations = face_recognition.face_locations(image, model="cnn")[0]
+			else:
+				face_locations = face_recognition.face_locations(image)[0]
+
+			face_encoding = face_recognition.face_encodings(image)[0]
+			for k in users.keys():
+				if users[k]['id'] == uid:
+					users[k]['image'] = base64_image.decode()
+					users[k]['face_locations'] = face_locations
+					users[k]['face_encoding'] = list(face_encoding)
+
+			with open('modules/databases/users.json', 'w') as f:
+				users = json.dump(users, f, indent=4)
+
+			os.remove(full_path)