Updated links in README + Added train_classifier.py & few improvements

codeglitchz · codeglitchz · commit 55e40d172aa6 · 2020-09-21T05:45:54.000+05:30
diff --git a/README.md b/README.md
@@ -9,7 +9,7 @@
 </p>
 
 > A simple, modern and scalable facial recognition based attendance system 
-> built with Python back-end & Angular front-end
+> built with Python back-end & Angular front-end.
 
 ![UI](sample/ui.gif)
 
@@ -25,14 +25,13 @@
 
 ## Prerequisites
 
-* Python v3.7+
-* Miniconda3 (optional) (recommended)
-* CMake
-* Visual Studio Build Tools
-* Nvidia CUDA (optional - for nvidia gpus)
-* Node.js LTS v12.8.0+ (npm v6.14.4+)
-* Angular CLI v9.1.8+
 * Windows or Linux (macOS not officially supported, but might work)
+* Nvidia CUDA (optional - for nvidia gpus)
+* [CMake](https://cmake.org/download/)
+* [Visual Studio Build Tools](https://visualstudio.microsoft.com/downloads/#build-tools-for-visual-studio-2019)
+* [Miniconda](https://docs.conda.io/en/latest/miniconda.html) for Python v3.7+
+* [Node.js LTS](https://nodejs.org/en/) v12.8.0+ (npm v6.14.4+)
+* [Angular CLI](https://cli.angular.io/) v9.1.8+
 
 ## Installation
 
@@ -94,21 +93,10 @@ $ cd frontend
 $ ng serve -o
 ```
 
-## Authors
-
-👤 [**@codeglitchz**](https://github.com/codeglitchz)
-
-👤 [**@codewhizz**](https://github.com/codewhizz)
-
 ## 🤝 Contributing
 
-Contributions, issues and feature requests are welcome!<br />Feel free to check [issues page](https://github.com/codeglitchz/attendance-system/issues). You can also take a look at the [contributing guide](https://github.com/codeglitchz/attendance-system/blob/master/CONTRIBUTING.md).
+Contributions, issues and feature requests are welcome!<br />Feel free to check [issues page](https://github.com/codeglitchz/attendance-system/issues).
 
 ## Show your support
 
-Give a ⭐️ if this project helped you!
-
-## 📝 License
-
-Copyright © 2020 [CodeGlitchz](https://github.com/codeglitchz). <br />
-This project is [MIT](https://github.com/codeglitchz/attendance-system/blob/master/LICENSE) licensed.
+Give a ⭐️ if this project helped you!
diff --git a/backend/run_cli.py b/backend/run_cli.py
@@ -3,6 +3,7 @@
 import pyfiglet
 
 from src.db import engine
+from src.libs.train_classifier import TrainClassifier
 from src.models import Base
 from src.settings import VIDEO_SOURCE
 from src.libs.cli_utils import CliAppUtils
@@ -53,7 +54,7 @@ def main_menu():
                 face_util.detect_n_capture()
                 break
             elif choice == 3:
-                face_util.train_classifier()
+                TrainClassifier.train()
                 break
             elif choice == 4:
                 face_util.recognize_n_attendance()
diff --git a/backend/src/app.py b/backend/src/app.py
@@ -8,7 +8,7 @@
 from src.libs.image_helper import IMAGE_SET
 from src.resources.dashboard import Dashboard
 from src.resources.teacher import Teacher, TeacherRegister, TeacherLogin
-from src.resources.student import StudentList, StudentAdd, StudentCapture, StudentDelete, TrainClassifier
+from src.resources.student import StudentList, StudentAdd, StudentCapture, StudentDelete
 from src.resources.attendance import AttendanceList
 from src.resources.video_feed import (
     VideoFeedList, VideoFeedAdd, VideoFeed, VideoFeedPreview, VideoFeedStop, VideoFeedStart, VideoFeedDelete
@@ -51,7 +51,6 @@ def handle_marshmallow_validation(err):
 api.add_resource(StudentAdd, "/students/add")
 api.add_resource(StudentCapture, "/students/capture/<int:student_id>")
 api.add_resource(StudentDelete, "/students/delete/<int:student_id>")
-api.add_resource(TrainClassifier, "/students/train")
 
 # /attendance
 api.add_resource(AttendanceList, "/attendance")
diff --git a/backend/src/libs/cli_utils.py b/backend/src/libs/cli_utils.py
@@ -4,13 +4,12 @@
 from datetime import date as dt
 
 import cv2
-import numpy as np
 import face_recognition
 
 from src.models import StudentModel, AttendanceModel
 from src.settings import (
     DATASET_PATH,
-    HAAR_CASCADE_PATH, PROTOTXT_PATH, CAFFEMODEL_PATH,
+    HAAR_CASCADE_PATH,
     DLIB_MODEL, DLIB_TOLERANCE, ENCODINGS_FILE
 )
 
@@ -93,66 +92,6 @@ def detect_n_capture(self):
         cap.release()
         cv2.destroyAllWindows()
 
-    @classmethod
-    def train_classifier(cls):
-        """
-        Train KNN Classifier by storing results in `files/encodings.pickle` file
-        """
-        # TODO: Store encodings in SQL database rather than `files/encodings.pickle` file
-        try:
-            print("[INFO] loading encodings...")
-            data = pickle.loads(open(ENCODINGS_FILE, "rb").read())
-            # initialize the list of known encodings and known names
-            known_encodings = data["encodings"]
-            known_ids = data["ids"]
-        except FileNotFoundError:
-            # initialize the list of known encodings and known names
-            known_encodings = []
-            known_ids = []
-
-        # get single unique ids by converting into set
-        # for each _id convert it into int
-        unique_ids = [int(_id) for _id in set(known_ids)]
-
-        # get all id_paths and join the path of the parent folder to each id_path
-        id_paths = [os.path.join(DATASET_PATH, f) for f in os.listdir(DATASET_PATH)]
-        # print(">>> ID paths:", id_paths)
-
-        # now looping through all the id_paths and loading the images in that id_path
-        for id_path in id_paths:
-            # getting the ID from the image
-            _id = int(os.path.split(id_path)[1])
-            if _id in unique_ids:
-                continue
-            # grab the paths to the input images of that ID
-            image_paths = [os.path.join(id_path, f) for f in os.listdir(id_path)]
-            for i, image_path in enumerate(image_paths):
-                print(f"[INFO] ID: {_id}, processing image {i + 1}/{len(image_paths)}")
-                # load the input image and convert it from RGB (OpenCV ordering)
-                # to dlib ordering (RGB)
-                image = cv2.imread(image_path)
-                rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
-
-                # detect the (x, y)-coordinates of the bounding boxes
-                # corresponding to each face in the input frame, then compute
-                # the facial embeddings for each face
-                boxes = face_recognition.face_locations(rgb, model=DLIB_MODEL)
-                # compute the facial embedding for the face
-                encodings = face_recognition.face_encodings(rgb, boxes)
-                # loop over the encodings
-                for encoding in encodings:
-                    # add each encoding + name to our set of known names and
-                    # encodings
-                    known_encodings.append(encoding)
-                    known_ids.append(_id)
-
-        # dump the facial encodings + names to disk
-        print("[INFO] serializing encodings...")
-        data = {"encodings": known_encodings, "ids": known_ids}
-        f = open(ENCODINGS_FILE, "wb")
-        f.write(pickle.dumps(data))
-        f.close()
-
     def recognize_n_attendance(self):
         print("[INFO] loading encodings...")
         data = pickle.loads(open(ENCODINGS_FILE, "rb").read())
diff --git a/backend/src/libs/train_classifier.py b/backend/src/libs/train_classifier.py
@@ -0,0 +1,76 @@
+import os
+import pickle
+
+import cv2
+import face_recognition
+
+from src.settings import DATASET_PATH, ENCODINGS_FILE, DLIB_MODEL
+
+
+class TrainClassifier:
+    """Train KNN Classifier by storing results in `files/encodings.pickle` file"""
+    # TODO: Store encodings in SQL database rather than `files/encodings.pickle` file
+    @classmethod
+    def train(cls):
+        try:
+            print("[INFO] loading encodings...")
+            with open(ENCODINGS_FILE, "rb") as ef:
+                data = pickle.loads(ef.read())
+            # initialize the list of known encodings and known names
+            known_encodings = data["encodings"]
+            known_ids = data["ids"]
+        except FileNotFoundError:
+            # initialize the list of known encodings and known names
+            known_encodings = []
+            known_ids = []
+
+        # get single unique ids by converting into set
+        # for each _id convert it into int
+        unique_ids = [int(_id) for _id in set(known_ids)]
+
+        # get all id_paths and join the path of the parent folder to each id_path
+        id_paths = [os.path.join(DATASET_PATH, f) for f in os.listdir(DATASET_PATH)]
+        # print(">>> ID paths:", id_paths)
+
+        # now looping through all the id_paths and loading the images in that id_path
+        for id_path in id_paths:
+            # getting the ID from the image
+            _id = int(os.path.split(id_path)[1])
+            if _id in unique_ids:
+                continue
+            # grab the paths to the input images of that ID
+            image_paths = [os.path.join(id_path, f) for f in os.listdir(id_path)]
+            for i, image_path in enumerate(image_paths):
+                print(f"[INFO] ID: {_id}, processing image {i + 1}/{len(image_paths)}")
+                # load the input image and convert it from RGB (OpenCV ordering)
+                # to dlib ordering (RGB)
+                image = cv2.imread(image_path)
+                try:
+                    rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
+                except cv2.error:
+                    # delete image that cannot be processed
+                    try:
+                        os.remove(image_path)
+                    except (FileNotFoundError, PermissionError):
+                        pass
+                    continue
+
+                # detect the (x, y)-coordinates of the bounding boxes
+                # corresponding to each face in the input frame, then compute
+                # the facial embeddings for each face
+                boxes = face_recognition.face_locations(rgb, model=DLIB_MODEL)
+                # compute the facial embedding for the face
+                encodings = face_recognition.face_encodings(rgb, boxes)
+                # loop over the encodings
+                for encoding in encodings:
+                    # add each encoding + name to our set of known names and
+                    # encodings
+                    known_encodings.append(encoding)
+                    known_ids.append(_id)
+
+        # dump the facial encodings + names to disk
+        print("[INFO] serializing encodings...")
+        data = {"encodings": known_encodings, "ids": known_ids}
+        f = open(ENCODINGS_FILE, "wb")
+        f.write(pickle.dumps(data))
+        f.close()
diff --git a/backend/src/resources/student.py b/backend/src/resources/student.py
@@ -2,19 +2,18 @@
 import shutil
 import pickle
 
-import cv2
-import face_recognition
 from flask import request
 from flask_restful import Resource
 from flask_uploads import UploadNotAllowed
 from flask_jwt_extended import jwt_required
 
 from src.db import Session
 from src.libs import image_helper
+from src.libs.train_classifier import TrainClassifier
 from src.libs.strings import gettext
 from src.models import StudentModel
 from src.schemas import StudentSchema, ImageSchema
-from src.settings import DATASET_PATH, ENCODINGS_FILE, DLIB_MODEL
+from src.settings import DATASET_PATH, ENCODINGS_FILE
 
 
 student_schema = StudentSchema()
@@ -101,68 +100,10 @@ def post(cls, student_id: int):
             save_as_filename = filename + extension
             image_path = image_helper.save_image(data["image"], folder=folder, name=save_as_filename)
             basename = image_helper.get_basename(image_path)
-            return {"message": gettext("image_uploaded").format(basename)}, 201
         except UploadNotAllowed:
             extension = image_helper.get_extension(data["image"])
             return {"message": gettext("image_illegal_extension").format(extension)}, 400
 
-
-class TrainClassifier(Resource):
-    """Train KNN Classifier by storing results in `files/encodings.pickle` file"""
-    # TODO: Store encodings in SQL database rather than `files/encodings.pickle` file
-    @classmethod
-    @jwt_required
-    def get(cls):
-        try:
-            print("[INFO] loading encodings...")
-            data = pickle.loads(open(ENCODINGS_FILE, "rb").read())
-            # initialize the list of known encodings and known names
-            known_encodings = data["encodings"]
-            known_ids = data["ids"]
-        except FileNotFoundError:
-            # initialize the list of known encodings and known names
-            known_encodings = []
-            known_ids = []
-
-        # get single unique ids by converting into set
-        # for each _id convert it into int
-        unique_ids = [int(_id) for _id in set(known_ids)]
-
-        # get all id_paths and join the path of the parent folder to each id_path
-        id_paths = [os.path.join(DATASET_PATH, f) for f in os.listdir(DATASET_PATH)]
-        # print(">>> ID paths:", id_paths)
-
-        # now looping through all the id_paths and loading the images in that id_path
-        for id_path in id_paths:
-            # getting the ID from the image
-            _id = int(os.path.split(id_path)[1])
-            if _id in unique_ids:
-                continue
-            # grab the paths to the input images of that ID
-            image_paths = [os.path.join(id_path, f) for f in os.listdir(id_path)]
-            for i, image_path in enumerate(image_paths):
-                print(f"[INFO] ID: {_id}, processing image {i + 1}/{len(image_paths)}")
-                # load the input image and convert it from RGB (OpenCV ordering)
-                # to dlib ordering (RGB)
-                image = cv2.imread(image_path)
-                rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
-
-                # detect the (x, y)-coordinates of the bounding boxes
-                # corresponding to each face in the input frame, then compute
-                # the facial embeddings for each face
-                boxes = face_recognition.face_locations(rgb, model=DLIB_MODEL)
-                # compute the facial embedding for the face
-                encodings = face_recognition.face_encodings(rgb, boxes)
-                # loop over the encodings
-                for encoding in encodings:
-                    # add each encoding + name to our set of known names and
-                    # encodings
-                    known_encodings.append(encoding)
-                    known_ids.append(_id)
-
-        # dump the facial encodings + names to disk
-        print("[INFO] serializing encodings...")
-        data = {"encodings": known_encodings, "ids": known_ids}
-        f = open(ENCODINGS_FILE, "wb")
-        f.write(pickle.dumps(data))
-        f.close()
+        # train images when submitted successfully
+        TrainClassifier.train()
+        return {"message": gettext("image_uploaded").format(basename)}, 201
diff --git a/frontend/src/app/components/video-feed-list/video-feed-list.component.html b/frontend/src/app/components/video-feed-list/video-feed-list.component.html
@@ -53,10 +53,9 @@
       <th scope="row">{{ feed.id }}</th>
       <td>
         <div *ngIf="!feed.is_active">
-          <fa-icon [icon]="circleIcon" style="color:red" size="xs"></fa-icon>&nbsp;
           Inactive
         </div>
-        <div *ngIf="feed.is_active" style="color:green">
+        <div *ngIf="feed.is_active" style="color:red">
           <fa-icon [icon]="circleIcon" size="xs"></fa-icon>&nbsp;
           <a class="nav-link" style="padding:0rem;display:inline" [routerLink]="generatePreviewUrl(feed.id)">Live</a>
         </div>
