Related
I am loading several images will go over my face and I am having difficulty getting the image to go over the square for face created. I have looked at a many resources , but for some reason I am receiving an error when attempting to follow their method.
Every time I do so , I receive an error
ValueError: could not broadcast input array from shape (334,334,3) into shape (234,234,3)
I think the images might be too large, however I tried to resize them to see if they will fit to no avail.
here is my code:
import cv2
import sys
import logging as log
import datetime as dt
from time import sleep
import os
import random
from timeit import default_timer as timer
cascPath = "haarcascade_frontalface_default.xml"
faceCascade = cv2.CascadeClassifier(cascPath)
#log.basicConfig(filename='webcam.log',level=log.INFO)
video_capture = cv2.VideoCapture(0)
anterior = 0
#s_img = cv2.imread("my.jpg")
increment = 0
for filename in os.listdir("Faces/"):
if filename.endswith(".png"):
FullFile = (os.path.join("Faces/", filename))
#ret, frame = video_capture.read()
frame = cv2.imread(FullFile, cv2.IMREAD_UNCHANGED)
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
faces = faceCascade.detectMultiScale( gray,scaleFactor=1.1, minNeighbors=5, minSize=(30, 30) )
edges = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 9, 9)
for (x, y, w, h) in faces:
roi_color = frame[y:( y ) + ( h ), x:x + w]
status = cv2.imwrite('export/faces_detected'+ str( increment ) +'.png', roi_color)
increment = increment + 1
else:
continue
masks = []
for filename in os.listdir("export/"):
if filename.endswith(".png"):
FullFile = (os.path.join("export/", filename))
s_img = cv2.imread(FullFile)
masks.append(s_img)
Start = timer()
End = timer()
MasksSize = len(masks)
nrand = random.randint(0, MasksSize -1 )
increment = 0
while True:
if not video_capture.isOpened():
print('Unable to load camera.')
sleep(5)
pass
# Capture frame-by-frame
ret, frame = video_capture.read()
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
faces = faceCascade.detectMultiScale(
gray,
scaleFactor=1.1,
minNeighbors=5,
minSize=(30, 30)
)
edges = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 9, 9)
# Draw a rectangle around the faces
for (x, y, w, h) in faces:
if (End - Start) > 3:
Start = timer()
End = timer()
nrand = random.randint(0, MasksSize -1 )
# -75 and +20 added to fit my face
cv2.rectangle(frame, (x, y - 75), (x+w, y+h+20), (0, 255, 0), 2)
s_img = masks[nrand]
increment = increment + 1
#maskresize = cv2.resize(s_img, (150, 150))
#frame[y:y+s_img.shape[0] , x:x+s_img.shape[1]] = s_img # problem occurs here with
# ValueError: could not broadcast input array from shape (334,334,3) into shape (234,234,3)
# I assume I am inserting somethign too big?
End = timer()
if anterior != len(faces):
anterior = len(faces)
#log.info("faces: "+str(len(faces))+" at "+str(dt.datetime.now()))
# Display the resulting frame
cv2.imshow('Video', frame)
#cv2.imshow('Video', cartoon)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
# Display the resulting frame
cv2.imshow('Video', frame)
# When everything is done, release the capture
video_capture.release()
cv2.destroyAllWindows()
In the following line,
frame[y:y+s_img.shape[0] , x:x+s_img.shape[1]] = s_img
you are trying to attempt to assign s_img to frame[y:y+s_img.shape[0] , x:x+s_img.shape[1]] which are of different shapes.
You can check the shapes of the two by printing the shape (it will be the same as the shapes mentioned in the error).
Try reshaping s_img to the same shape and then try to assign.
Refer to this link:https://www.geeksforgeeks.org/image-resizing-using-opencv-python/
I used this function to resize the image to scale.
def image_resize(image, width = None, height = None, inter = cv2.INTER_AREA):
# initialize the dimensions of the image to be resized and
# grab the image size
dim = None
(h, w) = image.shape[:2]
# if both the width and height are None, then return the
# original image
if width is None and height is None:
return image
# check to see if the width is None
if width is None:
# calculate the ratio of the height and construct the
# dimensions
r = height / float(h)
dim = (int(w * r), height)
# otherwise, the height is None
else:
# calculate the ratio of the width and construct the
# dimensions
r = width / float(w)
dim = (width, int(h * r))
# resize the image
resized = cv2.resize(image, dim, interpolation = inter)
# return the resized image
return resized
Then later on called
r= image_resize(s_img, height = h, width=w)
frame[y:y+r.shape[0] , x:x+r.shape[1]] = r
Answer taken from here too:
Resize an image without distortion OpenCV
I get the error below when running the following code:
import cv2, sys, numpy, os
haar_file = 'haarcascade_frontalface_default.xml'
datasets = 'datasets'
print('Recognizing Face Please Be in sufficient Lights...')
(images, lables, names, id) = ([], [], {}, 0)
for (subdirs, dirs, files) in os.walk(datasets):
for subdir in dirs:
names[id] = subdir
subjectpath = os.path.join(datasets, subdir)
for filename in os.listdir(subjectpath):
path = subjectpath + '/' + filename
lable = id
images.append(cv2.imread(path))
lables.append(int(lable))
id += 1
(width, height) = (130, 100)
(images, lables) = [numpy.array(lis) for lis in [images, lables]]
model = cv2.face.LBPHFaceRecognizer_create()
model.train(images, lables) # error comes here
face_cascade = cv2.CascadeClassifier(haar_file)
webcam = cv2.VideoCapture(0)
while True:
(_, im) = webcam.read()
gray = cv2.cvtColor(im, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray, 1.3, 5)
for (x, y, w, h) in faces:
cv2.rectangle(im, (x, y), (x + w, y + h), (255, 0, 0), 2)
face = gray[y:y + h, x:x + w]
face_resize = cv2.resize(face, (width, height))
prediction = model.predict(face_resize)
cv2.rectangle(im, (x, y), (x + w, y + h), (0, 255, 0), 3)
if prediction[1]<500:
cv2.putText(im, '% s' %
(names[prediction[0]]), (x-10, y-10),
cv2.FONT_HERSHEY_PLAIN, 1, (0, 255, 0))
else:
cv2.putText(im, 'not recognized',
(x-10, y-10), cv2.FONT_HERSHEY_PLAIN, 1, (0, 255, 0))
cv2.imshow('OpenCV', im)
key = cv2.waitKey(10)
if key == 27:
break
cv2.destroyAllWindows()
the error that pops up is:
Traceback (most recent call last):
File "Y:\vigyantram\AI-20200807T104521Z-001\AI\img processing1\face_recognize.py", line 19, in <module>
model.train(images, lables)
cv2.error: OpenCV(4.3.0) C:\projects\opencv-python\opencv_contrib\modules\face\src\lbph_faces.cpp:265: error: (-213:The function/feature is not implemented) Using Original Local Binary Patterns for feature extraction only works on single-channel images (given 16). Please pass the image data as a grayscale image! in function 'cv::face::elbp'?
Thank you in advance.
This may solve the problem. Use size(width and height) multiple of 8 or 16 if it doesn't work let me know u need to convert both test and train images to grey try that also it will work.
import cv2, sys, numpy, os
haar_file = 'haarcascade_frontalface_default.xml'
datasets = 'datasets'
print('Recognizing Face Please Be in sufficient Lights...')
(images, lables, names, id) = ([], [], {}, 0)
for (subdirs, dirs, files) in os.walk(datasets):
for subdir in dirs:
names[id] = subdir
subjectpath = os.path.join(datasets, subdir)
for filename in os.listdir(subjectpath):
path = subjectpath + '/' + filename
lable = id
images.append(cv2.imread(path))
lables.append(int(lable))
id += 1
(width, height) = (200, 200) #here 200 is multiple of 8
(images, lables) = [numpy.array(lis) for lis in [images, lables]]
model = cv2.face.LBPHFaceRecognizer_create()
model.train(images, lables) #error comes here
face_cascade = cv2.CascadeClassifier(haar_file)
webcam = cv2.VideoCapture(0)
while True:
(_, im) = webcam.read()
gray = cv2.cvtColor(im, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray, 1.3, 5)
for (x, y, w, h) in faces:
cv2.rectangle(im, (x, y), (x + w, y + h), (255, 0, 0), 2)
face = gray[y:y + h, x:x + w]
face_resize = cv2.resize(face, (width, height))
prediction = model.predict(face_resize)
cv2.rectangle(im, (x, y), (x + w, y + h), (0, 255, 0), 3)
if prediction[1]<500:
cv2.putText(im, '% s' %
(names[prediction[0]]), (x-10, y-10),
cv2.FONT_HERSHEY_PLAIN, 1, (0, 255, 0))
else:
cv2.putText(im, 'not recognized',
(x-10, y-10), cv2.FONT_HERSHEY_PLAIN, 1, (0, 255, 0))
cv2.imshow('OpenCV', im)
key = cv2.waitKey(10)
if key == 27:
break
cv2.destroyAllWindows()
i got to know what to do.
images.append(cv2.imread(path,0))
The '0' had to be added.
This is my Code of Mask Detection using YOLOV3 weights created by me. Whenever I run my Program, I experience a delay in my output Video of detection. This is the code please have a look.
import cv2
import numpy as np
net = cv2.dnn.readNet("yolov3_custom_final.weights", "yolov3_custom.cfg")
with open("obj.name", "r") as f:
classes = f.read().splitlines()
cap = cv2.VideoCapture(0 + cv2.CAP_DSHOW)
while True:
ret, img = cap.read()
height, weight, _ = img.shape
blob = cv2.dnn.blobFromImage(img, 1 / 255, (416, 416), (0, 0, 0), swapRB=True, crop=False)
net.setInput(blob)
output = net.getUnconnectedOutLayersNames()
layers = net.forward(output)
box = []
confidences = []
class_ids = []
for out in layers:
for detection in out:
scores = detection[5:]
class_id = np.argmax(scores)
confidence = scores[class_id]
if confidence > 0.3:
centre_x = int(detection[0] * weight)
centre_y = int(detection[1] * height)
w = int(detection[2] * weight)
h = int(detection[3] * height)
x = int(centre_x - w / 2)
y = int(centre_y - h / 2)
box.append([x, y, w, h])
confidences.append(float(confidence))
class_ids.append(class_id)
indexes = np.array(cv2.dnn.NMSBoxes(box, confidences, 0.5, 0.4))
font = cv2.FONT_HERSHEY_PLAIN
colors = np.random.uniform(0, 255, size=(len(box), 3))
for i in indexes.flatten():
x, y, w, h = box[i]
label = str(classes[class_ids[i]])
confidence = str(round(confidences[i], 2))
color = colors[i]
cv2.rectangle(img, (x, y), (x + w, y + h), color, 2)
cv2.putText(img, label + "I" + confidence, (x, y + 20), font, 2, (255, 255, 255), 2)
cv2.imshow("Final", img)
if cv2.waitKey(1) & 0xff == ord("q"):
break
cap.release()
cv2.destroyAllWindows()
Can someone Please help me in this Issue or suggest a way to reduce the Lag in my Output videostream ?
As I have done some research over the Time, I have a found a Possible answer to this question. As I'm running my YOLO model in my local system which has no GPU, This is the factor that is causing a delay in the Output as it Processes a frame and takes another frame after completion.
Basically, I have a function in my view which when called renders the opencv webcam videocapture window. But the problem is that the window opens but behind my chrome browser window, i have to manually click on the icon in taskbar to open the window ahead of the browser. Simply, the Z-index of the Capture window should be greater than the browser window.
VIEW.PY
def face_recognition(request, eid):
size = 4
haar_file = 'C:\\Users\\Aayush\\ev_manage\\face_detector\\haarcascade_frontalface_default.xml'
datasets = 'C:\\Users\\Aayush\\ev_manage\\face_detector\\static\\face_detector\\datasets'
# Create a list of images and a list of corresponding names
(images, lables, names, id) = ([], [], {}, 0)
for (subdirs, dirs, files) in os.walk(datasets):
for subdir in dirs:
names[id] = subdir
subjectpath = os.path.join(datasets, subdir)
for filename in os.listdir(subjectpath):
path = subjectpath + '/' + filename
lable = id
images.append(cv2.imread(path, 0))
lables.append(int(lable))
id += 1
# (width, height) = (130, 100)
# Create a Numpy array from the two lists above
(images, lables) = [numpy.array(lis) for lis in [images, lables]]
model = cv2.face.LBPHFaceRecognizer_create()
model.train(images, lables)
# Part 2: Use fisherRecognizer on camera stream
face_cascade = cv2.CascadeClassifier(haar_file)
webcam = cv2.VideoCapture(0)
flag = 0
d_id = 1000
while True:
(_, im) = webcam.read()
gray = cv2.cvtColor(im, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray, 1.3, 5)
for (x, y, w, h) in faces:
cv2.rectangle(im, (x, y), (x + w, y + h), (255, 0, 0), 2)
face = gray[y:y + h, x:x + w]
# face_resize = cv2.resize(face, (width, height))
face_resize = cv2.resize(face, None, fx=0.5, fy=0.5)
# Try to recognize the face
prediction = model.predict(face_resize)
cv2.rectangle(im, (x, y), (x + w, y + h), (0, 255, 0), 3)
if prediction[1] < 110:
uid = Delegate.objects.get(dataset_id=names[prediction[0]])
cv2.putText(im, '% s % s' %
(uid.first_name, uid.last_name), (x - -35, y - 20),
cv2.FONT_HERSHEY_PLAIN, 1, (0, 255, 0))
else:
cv2.putText(im, 'Unable to Recognize',
(x - -35, y - 20), cv2.FONT_HERSHEY_PLAIN, 1, (0, 255, 0))
if prediction[1] < 110:
d_id = names[prediction[0]]
# break
cv2.imshow('Recognizing Face...', im)
key = cv2.waitKey(10)
if key == 13:
present_delegate(request, eid, d_id)
break
webcam.release()
cv2.destroyAllWindows()
func = delegate_det(request, d_id)
return render(request, "face_recognition.html", {"eid": eid})
So is there any predefined method in the OpenCV library to force open the capture webcam window in a maximized form?
getting an error in the last if statement. show an syntax error as break is outside loop. The code is unable to break the statement when q is pressed.The code is from the github sorry for not providing the github link. ooking forward for some simple solutions. some packages have been imported like cv2, numpy and pandas
first_frame = None
status_list = [None, None]
times = []
df = pandas.DataFrame(columns=["start", "End"])
video = cv2.VideoCapture(0)
while True:
check, frame = video.read()
status = 0
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
gray = cv2.GaussianBlur(gray, (21, 21), 0)
if first_frame is None:
first_frame = gray
continue
delta_frame = cv2.absdiff(first_frame, gray)
thresh_delta = cv2.threshold(delta_frame, 30, 255, cv2.THRESH_BINARY)[1]
thresh_delta = cv2.dilate(thresh_delta, None, iterations=0)
countours, hierarchy = cv2.findContours(
thresh_delta.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
for contour in countours:
if cv2.contourArea(contour) < 1000:
continue
status = 1
(x, y, w, h) = cv2.boundingRect(contour)
cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 3)
status_list.append(status)
status_list = status_list[-2:]
if status_list[-1] == 1 and status_list[-2] == 0:
times.append(datetime.now())
if status_list[-1] == 0 and status_list[-2] == 1:
times.append(datetime.now())
cv2.imshow("frame", frame)
cv2.imshow("capturing", gray)
cv2.imshow("delta", delta_frame)
cv2.imshow("thresh", thresh_delta)
for i in range(0, len(times), 2):
df = df.append({"start": times[i], "End": times[i + 1]}, ignore_index=true)
df.to_csv("Times.csv")
if cv2.waitKey(1) & 0xFF == ord('q'): # press q to quit
break
video.release()
cv2.destroyAllWindows()
if cv2.waitKey(1) & 0xFF == ord('q'): # press q to quit
break