Real-Time Face Mask Detector with Python, OpenCV, Keras

During pandemic COVID-19, WHO has made wearing masks compulsory to protect against this deadly virus. In this tutorial we will develop a machine learning project – Real-time Face Mask Detector with Python.

face mask detector project

Real-Time Face Mask Detector with Python

We will build a real-time system to detect whether the person on the webcam is wearing a mask or not. We will train the face mask detector model using Keras and OpenCV.

Stay updated with latest technology trends
Join DataFlair on Telegram!!

Download the Dataset

The dataset we are working on consists of 1376 images with 690 images containing images of people wearing masks and 686 images with people without masks.

Download the dataset: Face Mask Dataset

Download the Project Code

Before proceeding ahead, please download the project source code: Face Mask Detector Project

Install Jupyter Notebook

In this machine learning project for beginners, we will use Jupyter Notebook for the development. Let’s see steps for the installation and configuration of Jupyter Notebook.

Using pip python package manager you can install Jupyter notebook:

pip3 install notebook

And that’s it, you have installed jupyter notebook

After installing Jupyter notebook you can run the notebook server. To run the notebook, open terminal and type:

jupyter notebook

It will start the notebook server at http://localhost:8888

jupyter notebook

To create a new project click on the “new” tab on the right panel, it will generate a new .ipynb file.

Create a new file and write the code which you have downloaded

Let’s dive into the code for face mask detector project:

We are going to build this project in two parts. In the first part, we will write a python script using Keras to train face mask detector model. In the second part, we test the results in a real-time webcam using OpenCV.

Make a python file train.py to write the code for training the neural network on our dataset. Follow the steps:

1. Imports:

Import all the libraries and modules required.

from keras.optimizers import RMSprop
from keras.preprocessing.image import ImageDataGenerator
import cv2
from keras.models import Sequential
from keras.layers import Conv2D, Input, ZeroPadding2D, BatchNormalization, Activation, MaxPooling2D, Flatten, Dense,Dropout
from keras.models import Model, load_model
from keras.callbacks import TensorBoard, ModelCheckpoint
from sklearn.model_selection import train_test_split
from sklearn.metrics import f1_score
from sklearn.utils import shuffle
import imutils
import numpy as np

2. Build the neural network:

This convolution network consists of two pairs of Conv and MaxPool layers to extract features from the dataset. Which is then followed by a Flatten and Dropout layer to convert the data in 1D and ensure overfitting.

And then two Dense layers for classification.

model = Sequential([
    Conv2D(100, (3,3), activation='relu', input_shape=(150, 150, 3)),
    MaxPooling2D(2,2),
    
    Conv2D(100, (3,3), activation='relu'),
    MaxPooling2D(2,2),
    
    Flatten(),
    Dropout(0.5),
    Dense(50, activation='relu'),
    Dense(2, activation='softmax')
])
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['acc'])

3. Image Data Generation/Augmentation:

TRAINING_DIR = "./train"
train_datagen = ImageDataGenerator(rescale=1.0/255,
                                   rotation_range=40,
                                   width_shift_range=0.2,
                                   height_shift_range=0.2,
                                   shear_range=0.2,
                                   zoom_range=0.2,
                                   horizontal_flip=True,
                                   fill_mode='nearest')

train_generator = train_datagen.flow_from_directory(TRAINING_DIR, 
                                                    batch_size=10, 
                                                    target_size=(150, 150))
VALIDATION_DIR = "./test"
validation_datagen = ImageDataGenerator(rescale=1.0/255)

validation_generator = validation_datagen.flow_from_directory(VALIDATION_DIR, 
                                                         batch_size=10, 
                                                         target_size=(150, 150))

4. Initialize a callback checkpoint to keep saving best model after each epoch while training:

checkpoint = ModelCheckpoint('model2-{epoch:03d}.model',monitor='val_loss',verbose=0,save_best_only=True,mode='auto')

5. Train the model:

history = model.fit_generator(train_generator,
                              epochs=10,
                              validation_data=validation_generator,
                              callbacks=[checkpoint])

project code

 

Now we will test the results of face mask detector model using OpenCV.

Make a python file “test.py” and paste the below script.

import cv2
import numpy as np
from keras.models import load_model
model=load_model("./model-010.h5")

results={0:'without mask',1:'mask'}
GR_dict={0:(0,0,255),1:(0,255,0)}

rect_size = 4
cap = cv2.VideoCapture(0) 


haarcascade = cv2.CascadeClassifier('/home/user_name/.local/lib/python3.6/site-packages/cv2/data/haarcascade_frontalface_default.xml')

while True:
    (rval, im) = cap.read()
    im=cv2.flip(im,1,1) 

    
    rerect_size = cv2.resize(im, (im.shape[1] // rect_size, im.shape[0] // rect_size))
    faces = haarcascade.detectMultiScale(rerect_size)
    for f in faces:
        (x, y, w, h) = [v * rect_size for v in f] 
        
        face_img = im[y:y+h, x:x+w]
        rerect_sized=cv2.resize(face_img,(150,150))
        normalized=rerect_sized/255.0
        reshaped=np.reshape(normalized,(1,150,150,3))
        reshaped = np.vstack([reshaped])
        result=model.predict(reshaped)

        
        label=np.argmax(result,axis=1)[0]
      
        cv2.rectangle(im,(x,y),(x+w,y+h),GR_dict[label],2)
        cv2.rectangle(im,(x,y-40),(x+w,y),GR_dict[label],-1)
        cv2.putText(im, results[label], (x, y-10),cv2.FONT_HERSHEY_SIMPLEX,0.8,(255,255,255),2)

    cv2.imshow('LIVE',   im)
    key = cv2.waitKey(10)
    
    if key == 27: 
        break

cap.release()

cv2.destroyAllWindows()

Run the project and observe the model performance.

python3 test.py

face mask detector project

Summary

In this project, we have developed a deep learning model for face mask detection using Python, Keras, and OpenCV. We developed the face mask detector model for detecting whether person is wearing a mask or not. We have trained the model using Keras with network architecture. Training the model is the first part of this project and testing using webcam using OpenCV is the second part.

This is a nice project for beginners to implement their learnings and gain expertise.

We work very hard to provide you quality material
Could you take 15 seconds and share your happy experience on Google | Facebook

61 Responses

  1. Firdos says:

    What you have used in the saved model ?? Getting error there

  2. Yoninah Sharma says:

    Same here. Were you able to solve the error? Can you please tell how?

  3. Divya says:

    Check the quotes ‘model2-009.model’

  4. Sasidhar says:

    UserWarning: image file could not be identified because WEBP support not installed
    UnidentifiedImageError: cannot identify image file

    this is happening when I’m training the model
    I installed libwebp libraries restarted the system after installing, even after that I’m getting the same error

  5. TALHA PIRZADA says:

    can anyone tell me how to upload test and train files on jupyter notebook

  6. Yash Gaikwad says:

    I’ve managed the program to run, but as soon as it runs I see the camera open notification but the camera does not open up, system says camera is in use but the camera screen does not open up. please someone help me I’ve come so far in this project with endless errors and depression. Please give me solution why my camera screen is not showing. thank you

  7. Shiv says:

    OSError: SavedModel file does not exist at: ./model-010.h5/{saved_model.pbtxt|saved_model.pb}

    where is the model-010.h5 file??
    i’m gettng error .

  8. MD. Abdulla Al Mamun says:

    After 1 epochs complete it shows an error: PIL.UnidentifiedImageError: cannot identify image file

    I tried to fix this problem from StackOverflow but there has no solution. can anyone help to solve this problem?

  9. Barani says:

    How to change the test. py program for image format? I have need this please post the answer. 🙏

  10. sasikanth says:

    I am getting NO MODULE NAMED KERAS please give me solution anyone

  11. Victor says:

    can I try in python instead of Jupyter?. When I tried in python. it is just open and close it. kindly help me to fix it.

  12. debopriyo says:

    managed to run the code.But it is always with Mask thought it is without in real.How to resolve

  13. Papia says:

    run the code.But result is wrong.always with mask even no mask.How to resolve.

  14. dhananjay says:

    FileNotFoundError: [WinError 3] The system cannot find the path specified: ‘c://Users/user/Desktop/project 6th sem/face-mask-dataset/Dataset/.train’
    can anyone help!

  15. shubham wankhade says:

    TRAINING_DIR = “./train”
    train_datagen = ImageDataGenerator(rescale=1.0/255,
    rotation_range=40,
    width_shift_range=0.2,
    height_shift_range=0.2,
    shear_range=0.2,
    zoom_range=0.2,
    horizontal_flip=True,
    fill_mode=’nearest’)

    train_generator = train_datagen.flow_from_directory(TRAINING_DIR,
    batch_size=10,
    target_size=(150, 150))
    VALIDATION_DIR = “./test”
    validation_datagen = ImageDataGenerator(rescale=1.0/255)

    validation_generator = validation_datagen.flow_from_directory(VALIDATION_DIR,
    batch_size=10,
    target_size=(150, 150))

    • shubham wankhade says:

      FileNotFoundError Traceback (most recent call last)
      in
      9 fill_mode=’nearest’)
      10
      —> 11 train_generator = train_datagen.flow_from_directory(TRAINING_DIR,
      12 batch_size=10,
      13 target_size=(150, 150))

      c:\users\shubham\appdata\local\programs\python\python39\lib\site-packages\keras\preprocessing\image.py in flow_from_directory(self, directory, target_size, color_mode, classes, class_mode, batch_size, shuffle, seed, save_to_dir, save_prefix, save_format, follow_links, subset, interpolation)
      953 and `y` is a numpy array of corresponding labels.
      954 “””
      –> 955 return DirectoryIterator(
      956 directory,
      957 self,

      c:\users\shubham\appdata\local\programs\python\python39\lib\site-packages\keras\preprocessing\image.py in __init__(self, directory, image_data_generator, target_size, color_mode, classes, class_mode, batch_size, shuffle, seed, data_format, save_to_dir, save_prefix, save_format, follow_links, subset, interpolation, dtype)
      371 dtype = backend.floatx()
      372 kwargs[‘dtype’] = dtype
      –> 373 super(DirectoryIterator, self).__init__(
      374 directory, image_data_generator,
      375 target_size=target_size,

      c:\users\shubham\appdata\local\programs\python\python39\lib\site-packages\keras_preprocessing\image\directory_iterator.py in __init__(self, directory, image_data_generator, target_size, color_mode, classes, class_mode, batch_size, shuffle, seed, data_format, save_to_dir, save_prefix, save_format, follow_links, subset, interpolation, dtype)
      113 if not classes:
      114 classes = []
      –> 115 for subdir in sorted(os.listdir(directory)):
      116 if os.path.isdir(os.path.join(directory, subdir)):
      117 classes.append(subdir)

      FileNotFoundError: [WinError 3] The system cannot find the path specified: ‘./train’

  16. prathyusha says:

    nice, but how to notify people that they are not wearing mask.

  17. Nared Fuengverojsakul says:

    Epoch 1/10
    102/132 [======================>…….] – ETA: 25s – loss: 0.5733 – acc: 0.7665
    —————————————————————————
    UnknownError Traceback (most recent call last)
    in
    —-> 1 history = model.fit(train_generator, epochs=10, validation_data=validation_generator, callbacks=[checkpoint])

    ~\anaconda3\lib\site-packages\tensorflow\python\keras\engine\training.py in fit(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq, max_queue_size, workers, use_multiprocessing)
    1181 _r=1):
    1182 callbacks.on_train_batch_begin(step)
    -> 1183 tmp_logs = self.train_function(iterator)
    1184 if data_handler.should_sync:
    1185 context.async_wait()

    ~\anaconda3\lib\site-packages\tensorflow\python\eager\def_function.py in __call__(self, *args, **kwds)
    887
    888 with OptionalXlaContext(self._jit_compile):
    –> 889 result = self._call(*args, **kwds)
    890
    891 new_tracing_count = self.experimental_get_tracing_count()

    ~\anaconda3\lib\site-packages\tensorflow\python\eager\def_function.py in _call(self, *args, **kwds)
    915 # In this case we have created variables on the first call, so we run the
    916 # defunned version which is guaranteed to never create variables.
    –> 917 return self._stateless_fn(*args, **kwds) # pylint: disable=not-callable
    918 elif self._stateful_fn is not None:
    919 # Release the lock early so that multiple threads can perform the call

    ~\anaconda3\lib\site-packages\tensorflow\python\eager\function.py in __call__(self, *args, **kwargs)
    3021 (graph_function,
    3022 filtered_flat_args) = self._maybe_define_function(args, kwargs)
    -> 3023 return graph_function._call_flat(
    3024 filtered_flat_args, captured_inputs=graph_function.captured_inputs) # pylint: disable=protected-access
    3025

    ~\anaconda3\lib\site-packages\tensorflow\python\eager\function.py in _call_flat(self, args, captured_inputs, cancellation_manager)
    1958 and executing_eagerly):
    1959 # No tape is watching; skip to running the function.
    -> 1960 return self._build_call_outputs(self._inference_function.call(
    1961 ctx, args, cancellation_manager=cancellation_manager))
    1962 forward_backward = self._select_forward_and_backward_functions(

    ~\anaconda3\lib\site-packages\tensorflow\python\eager\function.py in call(self, ctx, args, cancellation_manager)
    589 with _InterpolateFunctionError(self):
    590 if cancellation_manager is None:
    –> 591 outputs = execute.execute(
    592 str(self.signature.name),
    593 num_outputs=self._num_outputs,

    ~\anaconda3\lib\site-packages\tensorflow\python\eager\execute.py in quick_execute(op_name, num_outputs, inputs, attrs, ctx, name)
    57 try:
    58 ctx.ensure_initialized()
    —> 59 tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,
    60 inputs, attrs, num_outputs)
    61 except core._NotOkStatusException as e:

    UnknownError: UnidentifiedImageError: cannot identify image file
    Traceback (most recent call last):

    File “C:\Users\tam05\anaconda3\lib\site-packages\tensorflow\python\ops\script_ops.py”, line 249, in __call__
    ret = func(*args)

    File “C:\Users\tam05\anaconda3\lib\site-packages\tensorflow\python\autograph\impl\api.py”, line 645, in wrapper
    return func(*args, **kwargs)

    File “C:\Users\tam05\anaconda3\lib\site-packages\tensorflow\python\data\ops\dataset_ops.py”, line 961, in generator_py_func
    values = next(generator_state.get_iterator(iterator_id))

    File “C:\Users\tam05\anaconda3\lib\site-packages\tensorflow\python\keras\engine\data_adapter.py”, line 837, in wrapped_generator
    for data in generator_fn():

    File “C:\Users\tam05\anaconda3\lib\site-packages\tensorflow\python\keras\engine\data_adapter.py”, line 963, in generator_fn
    yield x[i]

    File “C:\Users\tam05\anaconda3\lib\site-packages\keras_preprocessing\image\iterator.py”, line 65, in __getitem__
    return self._get_batches_of_transformed_samples(index_array)

    File “C:\Users\tam05\anaconda3\lib\site-packages\keras_preprocessing\image\iterator.py”, line 227, in _get_batches_of_transformed_samples
    img = load_img(filepaths[j],

    File “C:\Users\tam05\anaconda3\lib\site-packages\keras_preprocessing\image\utils.py”, line 114, in load_img
    img = pil_image.open(io.BytesIO(f.read()))

    File “C:\Users\tam05\anaconda3\lib\site-packages\PIL\Image.py”, line 2967, in open
    raise UnidentifiedImageError(

    PIL.UnidentifiedImageError: cannot identify image file

    [[{{node PyFunc}}]]
    [[IteratorGetNext]] [Op:__inference_train_function_850]

    Function call stack:
    train_function

  18. Piyush Saha says:

    for this line is showing rerect_size = cv2.rerect_size(im, (im.shape[1] // rect_size, im.shape[0] // rect_size))
    # Resize the image to speed up detection
    —> 21 mini = cv2.resize(im, (im.shape[1] // size, im.shape[0] // size))
    22
    23 # detect MultiScale / faces

    AttributeError: ‘NoneType’ object has no attribute ‘shape’
    can you help me to solve?

Leave a Reply

Your email address will not be published. Required fields are marked *

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.