CODEWITHSUNDEEP
pythonopencv3.0face-recognition
Muhammad Ilham
Muhammad Ilham

Reputation: 31

396: error: (-5) This LBPH model is not computed yet. Did you call the train method? in function cv::face::LBPH::predict

I am trying to build a face recognition program using Python 2.7.13 and OpenCV 3.3.0.

However, I receive this error when it tries to detect my face:

File "C:\GitProject\face_recognition\detector.py", line 20, in
 <module> Id = recognizer.predict(gray[y:y+h,x:x+w])
error: C:\projects\opencv-python\opencv_contrib\modules\face\src\lbph_faces.cpp:396: 
error: (-5) This LBPH model is not computed yet. Did you call the train method? in function cv::face::LBPH::predict

Here is my code:

import cv2 ,os
import numpy as np
from PIL import Image
import pickle

recognizer = cv2.face.LBPHFaceRecognizer_create()
recognizer.read('trainer/training_data.yml')
cascadePath = "haarcascade_frontalface_default.xml"
faceCascade = cv2.CascadeClassifier(cascadePath);


cam = cv2.VideoCapture(0)
font = cv2.FONT_HERSHEY_SIMPLEX
while True:
    ret, im =cam.read()
    gray=cv2.cvtColor(im,cv2.COLOR_BGR2GRAY)
    faces=faceCascade.detectMultiScale(gray, 1.2,5)
    for(x,y,w,h) in faces:
        cv2.rectangle(im,(x,y),(x+w,y+h),(225,0,0),2)
        Id = recognizer.predict(gray[y:y+h,x:x+w])
        if(conf<50):
            if(Id==1):
                Id="Anirban"
            elif(Id==2):
                Id="Sam"
        else:
            Id="Unknown"
        cv2.PutText(cv2.fromarray(im),str(Id), (x,y+h),font, 255)
    cv2.imshow('im',im) 
    if cv2.waitKey(10) & 0xFF==ord('q'):
        break

cam.release()
cv2.destroyAllWindows()

Upvotes: 3

Views: 8970

Answers (4)

Brian Ramirez
Brian Ramirez

Reputation: 1

After all, if the error still there, try to manually specify the full path of recognizer.write().

Example:

recognizer.write('/home/pi/Desktop/.../trainer/trainer.yml')

This worked for me.

Upvotes: 0

Kieran Lyons
Kieran Lyons

Reputation: 47

The issue is caused when loading the pre-trained .yml file. Even though you do not receive an error when loading the file, the recognizer does not properly register the the .yml file.

By changing how the training_data.yml file is generated is the key. In your training code change the recognizer.save() to recognizer.write().

I had this issue when trying to load the training data on my Raspberry Pi.

If any clarity is needed make sure to ask.

This method caused model.yml to be incorrectly loaded in my facialRecognition class

Ids , faces = getImageList(path)
recognizer.train(faces,np.array(Ids))
recognizer.save('model.yml')

Changing the following line solved the issue.

recognizer.write('model.yml')

Both methods generated the model.yml file as expected but

recognizer.read('model.yml')

only worked with write().

Upvotes: 1

chb
chb

Reputation: 1995

Welcome to Stack Overflow. The error you received is telling you that you need to call the train method in the parent class of LBPHFaceRecognizer before you ask it to do any recognition. Refer to the documentation here.

Upvotes: 0

aiyagaze
aiyagaze

Reputation: 884

Open your trainer, change the recognizer.save to recognizer.write. Don't forget to run the trainer again. It works for me.

Upvotes: 10

Related Questions