0

I know how to using image detection to find whether there is helmet or not in the picture, but I don't know how to find whether there is someone without wearing helmet, especially when there is another wearing helmet in the picture.

Would you please tell me how to solve the problem? Thanks in advance. Ryu

P-Gn
  • 23,115
  • 9
  • 87
  • 104
Sekiichi
  • 3
  • 2

1 Answers1

0

First, it depends on what kind of helmets you want to include. Some helmets cover the face and don't.

In case of helmets that cover the face, you can use face detector or can train the faces a class. If face is detected, it means he/she is not wearing helmet.

In case of helmets the cover the face, you can detect the face and crop the appropriate area above the detected face in the image to detect the helmet.

Akhilesh
  • 1,024
  • 1
  • 9
  • 26
  • Thank you Akhilesh. My case is helmet which do NOT cover the face. I think croping the appropriate area above the detected face is a good idea, but problem is how to recognize the area whether helmet or not. – Sekiichi Jul 13 '17 at 09:26
  • As you wrote in the question **I know how to using image detection to find whether there is helmet or not in the picture**, create a model with two classed. One class for helmet and another for random object or background. Pass the random image image in the random class. Hope it will work. – Akhilesh Jul 13 '17 at 14:21