i am working on Odroid and running a face detection using open cv python on it. but there is too much lag in the camera. i have tried a lot of things but couldn't remove the lag. please suggest how can i remove the lag. i want to detect faces from at least 15 feet for that i need a high resolution, but high resolution causes more lag. currently i am having 2 second lag between frames. if there are any suggestions please share.

i am working on Odroid and running a face detection using open cv python on it. but there is too much lag in the camera. i have tried a lot of things but couldn't remove the lag. please suggest how can i remove the lag. i want to detect faces from at least 15 feet for that i need a high resolution, but high resolution causes more lag. currently i am having 2 second lag between frames..... if there are any suggestions please share.

i have solved it ... and Thanks a lot for the help :)

Thank you Berak my problem is solved but now i am having another problem i was trying to track the face but the tracking window doesn't do that its just span to half of the display window, is there anything that you can help with ?

Below is my code and i am having this error : File "imgTrack.py", line 37, in <module> cv2.imshow('img2',img2) cv2.error: /home/irum/OpenCv/modules/highgui/src/window.cpp:269: error: (-215) size.width>0 && size.height>0 in function imshow

Whats there that i am missing?

import numpy as np
import cv2
from matplotlib import pyplot as plt
cap = cv2.VideoCapture(0)

# take first frame of the video
ret,frame = cap.read()

# setup initial location of window
r,h,c,w = 250,90,400,125  # simply hardcoded the values
track_window = (c,r,w,h)

# set up the ROI for tracking
roi = frame[r:r+h, c:c+w]
hsv_roi =  cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)
mask = cv2.inRange(hsv_roi, np.array((0., 60.,32.)), np.array((180.,255.,255.)))
roi_hist = cv2.calcHist([hsv_roi],[0],mask,[180],[0,180])
cv2.normalize(roi_hist,roi_hist,0,255,cv2.NORM_MINMAX)

# Setup the termination criteria, either 10 iteration or move by atleast 1 pt
term_crit = ( cv2.TERM_CRITERIA_EPS | cv2.TERM_CRITERIA_COUNT, 10, 1 )

while(1):
    ret ,frame = cap.read()

    if ret == True:
        hsv = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)
        dst = cv2.calcBackProject([hsv],[0],roi_hist,[0,180],1)

        # apply meanshift to get the new location
        ret, track_window = cv2.CamShift(dst, track_window, term_crit)

        # Draw it on image
        pts = cv2.cv.BoxPoints(ret)
        pts = np.int0(pts)
        img2 = cv2.polylines(frame,[pts],True, 255,2)
        cv2.imshow('img2',img2)


        k = cv2.waitKey(60) & 0xff
        if k == 27:
            break
        else:
            cv2.imwrite(chr(k)+".jpg",img2)

    else:
        break

cv2.destroyAllWindows()
cap.release()

i solved the problem by using sklearn.cluster.KMeans it is making clusters now. I used the piece of code in you answer and then just passed the images to KMeans. `#Convert Images to Float32 images = np.asarray(images, np.float32) N = len(images) images = images.reshape(N,-1)

using kmeans clustring having 5 clusters

kmeans = KMeans(n_clusters=5)

passing images to kmeans

kmeans.fit(images)

centroids = kmeans.cluster_centers_ labels = kmeans.labels_ ` its working... Thank you for the help.

yes determining k is also another problem so i thought for now i can use k=10 because i have not that much different people in dataset. secondly if there is any other technique that can be more useful i will be very happy to try.

i am having this error now. you can see the code above i have edited the code. Traceback (most recent call last): File "Kmeans2.py", line 39, in <module> ret,label,center=cv2.kmeans(images,2,None,criteria,10,cv2.KMEANS_RANDOM_CENTERS) TypeError: an integer is required

(images, lables) = [numpy.array(lis) for lis in [images, lables]] these are two numpy arrays, one has faces and other has lables. i need to cluster image data, i want to differentiate one face from another so i can get same faces in one cluster.

i am doing detection and recognition , any face which is detected and not recognized is saved in a dataset. after a day my algorithm start doing clustering on that dataset in order to put them in a database of known people and each cluster will have a specific name after that. So when i tried doing clustering using open cv KMeans i am having different errors. Because i think i am passing wrong samples MAY BE. Problem is that i am asking how can i pass the images numpy array to this cv2.KMeans function ?

i will use this clustered data for face recognition .... basically i am doing clustering for face-recognition

i have a dataset of images and i want to do clustering on it. I have read the openCV documentations of kmeans but i just do not get it properly. Below is my code and i have no idea how can i pass images to the kmeans() and how to send all the clusters into different folders.

import cv2, sys, numpy, os
import matplotlib.pyplot as plt
from matplotlib import style
style.use("ggplot")
from sklearn.cluster import KMeans


fn_dir2 = 'unknown'
path2='/home/irum/Desktop/Face-Recognition/thakarrecog/unknown/UNKNOWNS'

# Create a list of images and a list of corresponding names
(images, lables, names, id) = ([], [], {}, 0)

#reading images from dataset
for (subdirs, dirs, files) in os.walk(fn_dir2):
    for subdir in dirs:
        names[id] = subdir 
        subjectpath = os.path.join(fn_dir2, subdir) 
        for filename in os.listdir(subjectpath):
            path = subjectpath + '/' + filename
            lable = id
            images.append(cv2.imread(path, 0))
            lables.append(int(lable))

        id += 1
#converting images and lables to numpy arrays 
(images, lables) = [numpy.array(lis) for lis in [images, lables]]
print "length images", len(images)
print type(images)
print "length lables", len(lables)

criteria = (cv2.TERM_CRITERIA_EPS + cv2.TERM_CRITERIA_MAX_ITER, 10, 1.0)

images = np.asarray(images, np.float32)
N = len(images)
images = images.reshape(N,-1)
ret,label,center=cv2.kmeans(images,k,None,criteria,10,cv2.KMEANS_RANDOM_CENTERS)

now i am having this error Traceback (most recent call last): File "Kmeans2.py", line 39, in <module> ret,label,center=cv2.kmeans(images,2,None,criteria,10,cv2.KMEANS_RANDOM_CENTERS) TypeError: an integer is required

i have a working copy of the code and that is working properly just slow.... i need to make that fast

i am having a very slow fps on raspberry pi when i run this code without this part and do the recognition straight away if counter<20: frame_list.append(eq) counter +=1 print "frame_list", len(frame_list) # Try to recognize the face for i in range(len(eq)): eq_i = eq[i] i thought if i do recognition for limited faces every time it would be better. but i am having an error. point is i just want to make fps better.

thankyou :)... let me try this now

normally true positive is less than 500 and true negative is more than 700. between 500 and 700 sometimes it recognizes accurately but some times it confuses some faces with other faces. that is why i am not sure about the threshold value.

ok now i get it.... secondly can you tell me how much confidence level should i take to tell confidently that the person is recognized correctly.

i have attached the code so you can take a look.

ok can you tell me what actually is this id ? its value changes from 1 to 4 for some faces.

i am using openCV to detect faces and then i start recognizing them using fisherface recognizer. and i didn't post my code above i posted the result of the FaceRecognizer.predict function.

i want to put a condition that if there is a known face then it executes the part where i display name of recognized person on the rectangle. but if there is an unknown face or a face with bad confidence level it should display "unknown" . my algorithm is working but sometimes it displays wrong name on a persons face. is there any way we i can solve this issue ?

facerec.py

import cv2, sys, numpy, os import json size = 4 fn_haar = 'haarcascade_frontalface_default.xml' fn_dir = 'att_faces'

Part 1: Create fisherRecognizer

print('Training...')

Create a list of images and a list of corresponding names

(images, lables, names, id) = ([], [], {}, 0) for (subdirs, dirs, files) in os.walk(fn_dir): for subdir in dirs: names[id] = subdir subjectpath = os.path.join(fn_dir, subdir) for filename in os.listdir(subjectpath): path = subjectpath + '/' + filename lable = id images.append(cv2.imread(path, 0)) lables.append(int(lable)) id += 1 (im_width, im_height) = (112, 92)

Create a Numpy array from the two lists above

(images, lables) = [numpy.array(lis) for lis in [images, lables]]

OpenCV trains a model from the images

NOTE FOR OpenCV2: remove '.face'

model = cv2.createFisherFaceRecognizer() model.train(images, lables)

Part 2: Use fisherRecognizer on camera stream

haar_cascade = cv2.CascadeClassifier(fn_haar) webcam = cv2.VideoCapture(0) while True: (rval, frame) = webcam.read() frame=cv2.flip(frame,1,0) gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY) mini = cv2.resize(gray, (gray.shape[1] / size, gray.shape[0] / size)) faces = haar_cascade.detectMultiScale(mini) for i in range(len(faces)): face_i = faces[i] (x, y, w, h) = [v * size for v in face_i] face = gray[y:y + h, x:x + w] face_resize = cv2.resize(face, (im_width, im_height))

    # Try to recognize the face
    prediction  = model.predict(face_resize)


    cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 3)

    # Write the name of recognized face
    # [1]
result = {
            'face': {

             'distance': prediction,
             'coords': {
               'x': str(faces[0][0]),
               'y': str(faces[0][1]),
               'width': str(faces[0][2]),
               'height': str(faces[0][3])
                }
            }
          }
print result
if prediction > 0:
    if prediction[1]>0 and prediction[1]<500:

         result = {
            'face': {

             'distance': prediction[1],
             'coords': {
               'x': str(faces[0][0]),
               'y': str(faces[0][1]),
               'width': str(faces[0][2]),
               'height': str(faces[0][3])
                }
            }
          }

         dist = result['face']['distance']
         print "DIST" , dist
         print "prediction",prediction 
         cv2.putText(frame,
                 '%s - %.0f' % (names[prediction[0]],prediction[1]),
                 (x-10, y-10), cv2.FONT_HERSHEY_PLAIN,1,(0, 255, 0))
else:
     print "prediction",prediction 
     result = {
        'face': {

         'distance': prediction[0],
         'coords': {
           'x': str(faces[0][0]),
           'y': str(faces[0][1]),
           'width': str(faces[0][2]),
           'height': str(faces[0][3])
            }
        }
      }
     dist = result['face']['distance']
     print "DIST Unknown" , dist
     cv2.putText(frame,
          'Unknown',
          (x-10, y-10), cv2.FONT_HERSHEY_PLAIN,1,(0, 255, 0))
cv2.imshow('OpenCV', frame)
key = cv2.waitKey(10)
if key == 27:
    break

whats the difference between distance value and the value in prediction ? how can i make it optimize using distance. what one value should i set to determine recognized face.