neha01/facerecognition

Face Recognition using OpenCV in Python

Face Recognition

Face Recognition using OpenCV in Python

Youtube Tutorial

https://youtu.be/h21wMKGs0qs

Prerequisites

Numpy
OpenCV

Note: Please install opencv-contrib-python package instead of opencv-contrib as it contains the main modules and also contrib modules.

Installing

Install Numpy via anaconda: conda install numpy

Install OpenCV via anaconda: conda install -c menpo opencv

Running the tests

Run Tester.py script on commandline to train recognizer on training images and also predict test_img:

python tester.py

1.Place some test images in TestImages folder that you want to predict in tester.py file
2.Place Images for training the classifier in trainingImages folder.If you want to train clasifier to recognize multiple people then add each persons folder in separate label markes as 0,1,2,etc and then add their names along with labels in tester.py/videoTester.py file in 'name' variable.
3.To generate test images for training classifier use videoimg.py file.
4.To do test run via tester.py give the path of image in test_img variable
5.Use "videoTester.py" script for predicting faces realtime via your webcam.(But ensure that you run tester.py first since it generates training.yml file that is being used in "videoTester.py" script.

Acknowledgments

• https://www.superdatascience.com/opencv-face-recognition/
• https://pythonprogramming.net/haar-cascade-face-eye-detection-python-opencv-tutorial/

facerecognition's Issues

How to train Identify for more people?

How to train like, giving ID of person and getting dataset and training them.
Can we store, trained data of different people in same .yml file ?

AttributeError

module 'cv2.cv2' has no attribute 'face'

in line---> "face_recognizer=cv2.face.LBPHFaceRecognizer_create()"

Why do we have resizedimagesfolder here ?

This is not an issue but curious to know why do we have resizedimages folder. Code worked fine without that folder too.
But if I have to use this code on my own dataset, can I resize it first and run the trainer to save time ? Will the accuracy be same ?

Error on OpenCV4.1.1

cv2.error: OpenCV(4.1.1) /home/pi/opencv_contrib/modules/face/src/lbph_faces.cpp:365: error: (-210:Unsupported format or combination of formats) Labels must be given as integer (CV_32SC1). Expected 4, but was 6. in function 'train'.
Can you please help me ?

cv::CascadeClassifier::detectMultiScale

Traceback (most recent call last):
File "C:\Users\DELL\Desktop\openCV.py", line 18, in
faces=face_cascade.detectMultiScale(gray_img,scaleFactor=1.32,minNeighbors=5)
cv2.error: OpenCV(4.1.0) C:\projects\opencv-python\opencv\modules\objdetect\src\cascadedetect.cpp:1658: error: (-215:Assertion failed) !empty() in function 'cv::CascadeClassifier::detectMultiScale'

I don't know what to do in this phase!

opencv error

File "tester.py", line 9, in
faces_detected,gray_img=fr.faceDetection(test_img)
File "C:\Users\sunil\Desktop\ml_project\me\face_recognition.py", line 6, in faceDetection
gray_img=cv2.cvtColor(test_img,cv2.COLOR_BGR2GRAY)
cv2.error: OpenCV(4.1.0) C:\projects\opencv-python\opencv\modules\imgproc\src\color.cpp:182: error: (-215:Assertion failed) !_src.empty() in function 'cv::cvtColor'

this is the error i m getting.Can you tell me what to do in that

Is there a typo in line 9 in tester.py

faces_detected,gray_img=fr.faceDetection(test_img)

faces_detected.gray_img=fr.faceDetection(test_img)

Error in printing faces_detected

I am having this error on printing faces_detected after using faceDetection module -

OpenCV(3.4.2) c:\miniconda3\conda-bld\opencv-suite_1534379934306\work\modules\imgproc\src\color.hpp:253: error: (-215:Assertion failed) VScn::contains(scn) && VDcn::contains(dcn) && VDepth::contains(depth) in function 'cv::CvtHelper<struct cv::Set<3,4,-1>,struct cv::Set<1,-1,-1>,struct cv::Set<0,2,5>,2>::CvtHelper'

in the line -
gray_img = cv2.cvtColor(test_img,cv2.COLOR_BGR2GRAY)

how to train images present in DataBase?

how to train images present in DataBase?

Instead of training photos from a folder, how can we train images present in data base..?

Haar cascade

Can you define Haar cascades?

opencv error

File "tester.py", line 9, in
faces_detected,gray_img=fr.faceDetection(test_img)
File "C:\Users\sunil\Desktop\ml_project\me\face_recognition.py", line 6, in faceDetection
gray_img=cv2.cvtColor(test_img,cv2.COLOR_BGR2GRAY)
cv2.error: OpenCV(4.1.0) C:\projects\opencv-python\opencv\modules\imgproc\src\color.cpp:182: error: (-215:Assertion failed) !_src.empty() in function 'cv::cvtColor'

this is the error i m getting.Can you tell me what to do in that

Opencv version

Which version of opencv should I install ???