atulyakumar97/fire-and-gun-detection

Fire and Gun detection using yolov3 in videos as well as images. Training code, dataset and trained weight file available.

Fire and Gun Detection

How to use yolo.py:

usage: yolo.py [-h] [--webcam WEBCAM] [--play_video PLAY_VIDEO]
               [--image IMAGE] [--video_path VIDEO_PATH]
               [--image_path IMAGE_PATH] [--verbose VERBOSE]

optional arguments:
  -h, --help            show this help message and exit
  --webcam WEBCAM       True/False
  --play_video PLAY_VIDEO
                        Tue/False
  --image IMAGE         Tue/False
  --video_path VIDEO_PATH
                        Path of video file
  --image_path IMAGE_PATH
                        Path of image to detect objects
  --verbose VERBOSE     To print statements

Weights File Backup

If the GitLFS file is not accessible - Download Weights and keep inside the project folder.

Move inside the project folder and use the following command:

python yolo.py --play_video True --video_path videos/fire1.mp4

Dataset

Training done on google collab - Jupyter notebook

Demo: Youtube

Paper

Fire and Gun Violence based Anomaly Detection System Using Deep Neural Networks
Proceedings of the International conference on Electronics and Sustainable Communication Systems - ICESCS 2020
ISBN: 978-1-7281-4107-7
978-1-7281-4108-4/20/©2020 IEEE

fire-and-gun-detection's Issues

Your weight and config file can't support other type of code

Your weight and config file can't support other type of code.only support your code.

How to improve the detect speed?

Hi:
I noticed in your sample videos, there is no jitters or lags at all. But when i run your default code, it has a lot of jitters/lags. A 5 seconds video may need 15 seconds to finish. Looks like the net.forward part is taking too long. How to modify the code in order to get the same effect as you showed in your video? Many thanks!

def detect_objects(img, net, outputLayers):			
	blob = cv2.dnn.blobFromImage(img, scalefactor=0.00392, size=(320, 320), mean=(0, 0, 0), swapRB=True, crop=False)
	net.setInput(blob)
	outputs = net.forward(outputLayers) #this part is taking so long
	return blob, outputs

The yolov3.weights file is no longer available

Please provide a download link. Thank you!

Error

Great work you did there :)

I am getting an error while running this code, can you please help me.

Opening videos/fire1.mp4 ....
Traceback (most recent call last):
File "yolo.py", line 143, in
start_video(video_path)
File "yolo.py", line 116, in start_video
model, classes, colors, output_layers = load_yolo()
File "yolo.py", line 24, in load_yolo
output_layers = [layers_names[i[0]-1] for i in net.getUnconnectedOutLayers()]
File "yolo.py", line 24, in
output_layers = [layers_names[i[0]-1] for i in net.getUnconnectedOutLayers()]
IndexError: invalid index to scalar variable.

I would be really glad if you could help me in this.
Thank you :)