Real-time object detection using deep learning, Python, and OpenCV

In this video I demo real-time object detection using deep learning, Python, and OpenCV. The source code + tutorial can be found using the link above.

[DarkShadow]

Hi adrain big fan of ur works
I have to ask in videostream(0) in place of (0) if im putting some ip address like videostream('http..'), it is giving me like no shape error
What should i do to run ip camera in ur code. Plz answer my question
Thanks

[iamapilotstar]

how to add a voice feedback function to the code? like the object is detected and a voice feedback is given on the object that is detected?

[zhixianjin]

where is the txt file you typed at beginning in the cmd? Do we need to create that by myself? Thank you.

[jericreyechavez]

Hi Adrian, do have a tutorial on real-time measuring the height of a person using opencv?

[hitanshurami]

How to add more objects in models?like suppose I want it to detect mobile phone as well how do I do that?

[misbahshaikh]

On running the Source Code(Python file) which I received on my mail, I got an error stating ModuleNotFoundError: No module named 'imutils'

[alinamustafa]

Code plzzzz i have my fyp mid year presentation on 25th :-(

[afkniladri]

could you please help me to rectify the error !

blob = cv2.dnn.blobFromImage(cv2.resize(frame, (300, 300)), 0.007843, (300, 300), 127.5)
AttributeError: module 'cv2.dnn' has no attribute 'blobFromImage'

[techlappy]

usage: real_time_object_detection.py [-h] -p PROTOTXT -m MODEL [-c CONFIDENCE]
error: the following arguments are required: -p/--prototxt, -m/--model
how to fix this

[KandelDeepak]

I am using Picamera..and I got the error like....[INFO] loading model...
[INFO] starting video stream...
Traceback (most recent call last):
File "real_time_object_detection.py", line 48, in <module>
frame = imutils.resize(frame, width=400)
File "/home/pi/.virtualenvs/cv/lib/python3.5/site-packages/imutils/convenience.py", line 69, in resize
(h, w) = image.shape[:2]

[opemipodurodola]

Hello Adrian, do you use Tensorflow ?

[archanamuraleedharan]

Hii Adrian..I got the following error while running the code:
usage: new.py [-h] -i IMAGE -p PROTOTXT -m MODEL [-c CONFIDENCE]
new.py: error: argument -i/--image is required

[darrenmagturo]

can i add speaker to its feature?
for example the object was detected then the speaker would say what object is it?
is it possible to do?
thanks a lot !

[thomaschung]

Hello, Adrian, may I know if its possible yo use custom video stream frame for it? Other than webcam. Thank You. It works on Windows

[CodeBlazeKAT]

i got an error like this:
usage: real_time_object_detection.py [-h] -p PROTOTXT -m MODEL [-c CONFIDENCE]
error: argument -p/--prototxt is required

i'm not familiar with this stuff and i'm seriously stuck.Please help

[lokeshkumar]

Hello Adrian.. The link is not working!!

[kinalgoti]

hi Adrian.. It gives me error like

Trackback (most recent call last):
File line 5, in <module>
"execution_count": null,
NameError: name 'null' is not defined

can you help me to solve this error and how can i run it.?
thank you

[babypiro]

def detect(image):

gray=image


blob = cv2.dnn.blobFromImage(gray, 1.0, (300, 300), [104, 117, 123], False, False)
net.setInput(blob)
detections = net.forward()
bboxes = []
gray=cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)
frameWidth=image.shape[1]
frameHeight=image.shape[0]
for i in range(detections.shape[2]):
confidence = detections[0, 0, i, 2]
if confidence > 0.6:
x1 = int(detections[0, 0, i, 3] * frameWidth)
y1 = int(detections[0, 0, i, 4] * frameHeight)
x2 = int(detections[0, 0, i, 5] * frameWidth)
y2 = int(detections[0, 0, i, 6] * frameHeight)
cv2.rectangle(image, (x1, y1), (x2, y2), (255, 0, 0), 2)
try:
image1 = gray[y1:(y2), x1:(x2)]

img = cv2.resize(image1, (48, 48), interpolation = cv2.INTER_CUBIC) / 255.

prediction=model1.predict_proba(img.reshape(1, 48, 48, 1))



plot_images(img, 1)
plt.show()
print("Emotion Detected: -
emt=[prediction[0][0], prediction[0][1], prediction[0][2], prediction[0][3], prediction[0][4], prediction[0][5], prediction[0][6]]
indx=np.arange(len(emotion))
plt.bar(indx, emt)
plt.xticks(indx, emotion)
plt.show()
except:
during resize .Probably Cant detect any face")

Could anyone help me in getting this code run properly, here I am using the face detector DNN, the above code is running fine but it is creating 2 different bounding boxes one around the face and another in the side where is no face.Is thr some issue in the above somebody told me that in DNN Face detector I dont have to convert to grayscale....but my neural net accepts grayscale meaning 1 channel so for that reason I have converted.
Apart from that what is if confidence > 0.6:
x1 = int(detections[0, 0, i, 3] * frameWidth)
y1 = int(detections[0, 0, i, 4] * frameHeight)
x2 = int(detections[0, 0, i, 5] * frameWidth)
y2 = int(detections[0, 0, i, 6] * frameHeight)
cv2.rectangle(image, (x1, y1), (x2, y2), (255, 0, 0), 2)

part doing....?I am a bit new ..

[AFuller]

Not to sound stuffy, but unless you can bridge the deployment obstacles, the parlor tricks are meaningless. Virtualbox is will not run 24/7, VM"S are too big and Docker, well it's Docker. So the only way to use your opencv is with a dedicated linux "box".