Thief-Detection : Person Detection is Thief or Not (Based on taking up an unusual path) about thief_detection HOT 40 OPEN

Yes I have defined them. I will still again look onto it. Also I think the packages are not installed properly.

from thief_detection.

I want to work on this issue
#6

from thief_detection.

You can work on only 1 issue at a time so let me know of what you would choose first? @kkg2001

from thief_detection.

I want to work on issue#6

from thief_detection.

Ok, provide the details mentioned to get it assigned. @kkg2001

from thief_detection.

Name-Krishna Kant Gupta
Participant ID:499
Discord-krish1234#9508
Project:Person Detection is thief or not
github id:@kkg2001

from thief_detection.

Approach please. @kkg2001

from thief_detection.

Approach: First of all we will check the stranger's body movements and take into consideration his body language from the usual parameters. Secondly using opencv or any other technology relevant to this project we will check whether the person is nervous/curious or not keeping into consideration how normal people's face reactions. Time parameter like whether the stranger looks in a hurry or he is pretending to be in normal state and taking more time than usual so that nobody catches him. There are chances that he may not use the usual path to enter the house in that case the siren alarm should ring immediately and should confirm that he is certainly a thief. Tracing his finger prints or detecting his face could make work easier whether his history is there as a thief.

from thief_detection.

This would be a long thing for this issue. Nice ideas you have showcased. But we already created other issues for considerations mentioned. For this issue, we shall do it with basic one. If person is not entering through a usual path, we detect him as thief and alarm would ring immediately. @kkg2001

from thief_detection.

Thanks. Ok I will work on only the basic one when the thief is entering through an unusual path

from thief_detection.

Yes, once you finish this. You can take up an another issue related to it. Issue is assigned to @kkg2001

from thief_detection.

Yours is a basic and will be interlinked issue for many other issues. @kkg2001

from thief_detection.

Ok I will make sure I put up my best efforts for this issue and then I will move over to other issues (probably #7)

from thief_detection.

Great, you can proceed. All the best. @kkg2001

from thief_detection.

I am presuming that cctv cameras are installed on that path from where stranger will try to enter the house from an unusual path. I will share the code shortly and let me know the changes in it.
Thanks

from thief_detection.

Ok sure. Yeah, with cctv. Without cctv, you have any idea to add? I don't think owners would keep cctv on unsual path. For now, this works @kkg2001

from thief_detection.

If by some mechanism we can measure the randomness of motion of the stranger coming by unusual path then its possible(something like a sensor or if we can predict its motion history). For example coming via window pane or jumping off the fence will have different motions.

from thief_detection.

Yes, for this issue we shall proceed with this thing we had thought of. We can add this up later in issues for better proximity. @kkg2001

from thief_detection.

You mean we can proceed with this cctv thing what we actually presumed or something else?

from thief_detection.

It's the same what we had thought earlier. Proceeding with CCTV thing. @kkg2001

from thief_detection.

Yes, the same. I thought you would have started working. @kkg2001

from thief_detection.

I started back then but in the back of my mind was thinking about what if the cctv was not in the unusual path. Regarding the coding part I am facing some errors so I am fixing those.

from thief_detection.

Ok for now we will get this done and later add enhancements to it. What are the errors you're getting? @kkg2001

from thief_detection.

'error: unrecognsied elements' , 'an exception has occured , use %tb to see the full traceback'

from thief_detection.

*arguments not elements

from thief_detection.

The elements you have mentioned in arguments aren't recognizing. Have you defined them? @kkg2001

from thief_detection.

Ok, Understood libraries should be loaded for functionality to work. @kkg2001

from thief_detection.

Kindly update the progress. Its been more than 10days. It should be done by now, if you're stuck anywhere we are here to help out. @kkg2001

from thief_detection.

from thief_detection.

Because most of the work initiates from this issue. @kkg2001

from thief_detection.

from thief_detection.

Oh, if you could share me out the link @kkg2001

from thief_detection.

import cv2
import numpy as np

body_classifier = cv2.CascadeClassifier(cv2.data.haarcascades+'haarcascade_fullbody.xml')

cap = cv2.VideoCapture('Thief.avi')

while cap.isOpened():

ret, frame = cap.read()


gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

bodies = body_classifier.detectMultiScale(gray, 1.2, 3)


for (x,y,w,h) in bodies:
    cv2.rectangle(frame, (x, y), (x+w, y+h), (0, 255, 255), 2)
    cv2.imshow('Thief', frame)

if cv2.waitKey(): 
    break

cap.release()
cv2.destroyAllWindows()

from thief_detection.

This is the code which is giving no response

from thief_detection.

I made the required changes in the code and its working fine. Can I proceed towards merging the PR?

from thief_detection.

oops, sorry I was unwell. I couldn't check out your messages. Hope error is solved? Yes, if it's working you can proceed to do so
@kkg2001

from thief_detection.

I am stuck in this. What to do?

from thief_detection.

I guess your forked repo and main repo is not even. That's why it isn't dedicating the changes made. Make it even and then push it back @kkg2001

from thief_detection.

from thief_detection.

Yes @kkg2001

from thief_detection.