The document discusses the implementation of automatic weapon detection using convolutional neural networks (CNN) and faster R-CNN algorithms with two types of datasets, including pre-labelled and manually labelled images. The system aims to improve security through efficient detection, revealing that the SSD algorithm provides faster processing, while faster R-CNN offers higher accuracy. The conclusion highlights a trade-off between speed and accuracy, with potential for scalability using advanced hardware.

1. WEAPON DETECTION
USING
ARTIFICIAL INTELLIGENCE AND DEEP LEARNING
Under the Guidance of
Mr.CH.BALA KRISHNA Sir
- Asst.prof.CSE dept
1. J.Koteswarrao - 19C51A0564
2. K.Vijay kumar - 19C51A0571
3. V.Goutham - 19C51A0591
4. G.Ravi teja - 19C51A0561
Presented by -

2. ABSTRACT
 InThis we implements automatic weapon detection using a
Convolution Neural Network (CNN) and Faster RCNN(Region-
Convolutional Neural Network) algorithms. Proposed
implementation uses two types of datasets. One dataset, which
had pre-labelled images and the other one is a set of images,
which were labelled manually.

3. INTRODUCTION
• Security is always a main conern in every domain, due to a rise in
crime rate in a crowded event or suspicious lonely areas.
Abnormal detection and monitoring have major applications of
computer vision to tackle various problems. Due to growing
demand in the protection of safety, security and personal
properties, needs and deployment of video surveillance systems
an recognize and interpret the scene and anomaly events play a
vital role in intelligence monitoring.

4. EXISTING SYSTEM
• Existing methods used aTensorflow-based implementation of
the Overfeat network as an integrated network for detecting
and classifying weapons in images.

5. PROPOSED SYSTEM
• We Proposed a Convolution Neural Network (CNN) based
SSD(Single-Shot Detector) and Faster RCNN algorithms for
weapon detection.

6. ADVANTAGE
• These improvements allow SSD to match the Faster R-CNN’s
accuracy using lower resolution images, which further pushes
speed higher and lower cost.

7. DIS ADVANTAGE
• less efficiency

8. FLOW CHART

9. TECHNOLOGY
 Machine learning
 Deep learning
 Python packages

10. PROGRAMMING LANGUAGE AND
PACKAGES
• Python
• Numpy, pandas, keras, sklearn, tkintertable, matplotlib, pillow,
imutils.
• Tensorflow

11. SOFTWARE
 Python idel 3.7 version (or)
 Anaconda 3.7 ( or)
 Jupiter

12. HARDWARE
Operating system: windows, linux
Processor : minimum intel i3
Ram: minimum 4 gb
Harddisk : minimum 250gb

13. ALGORITHM MODELS
• CNN (Convolutional Neural Network)
A CNN is a kind of network architecture for deep learning algorithms and is
specifically used for image recognition and tasks that involve the processing of pixel
data.
• RCNN (Region-Based Convolutional Neural Network)
RCNN is a type of machine learning model that is used for computer vision
tasks, specifically for object detection.

14. WORKING MODULES
• Dataset upload
• Pre-processing data
• Extracting dataset
• Spliting dataset training and testing
• image processing
• Applying models

15. PROJECT IMPLEMENTATION
• Gathering the dataset from database
• Pre-processing the dataset and analysis dataset
• Splitting the datasets into training and testing in the ration 80%
and 20% and CNN and RCNN models to analysis the data
• Obtain the accuracy in prediction

16. INPUT
• The input is image files
OUTPUT
• The output is the accuracy in
prediction

17. INPUT:

18. OUTPUT:

19. CONCLUSION
• SSD and Faster RCNN algorithms are simulated for pre labeled and self-created image dataset
for weapon (gun) detection. Both the algorithms are efficient and give good results but their
application in real time is based on a tradeoff between speed and accuracy. In terms of speed,
SSD algorithm gives better speed with 0.736 s/frame.Whereas Faster RCNN gives speed
1.606s/frame, which is poor compared to SSD.With respect to accuracy, Faster RCNN gives
better accuracy of 84.6%.Whereas SSD gives an accuracy of 73.8%, which is poor compared to
faster RNN.SSD provided real time detection due to faster speed but Faster RCNN provided
superior accuracy. Further, it an be implemented for larger datasets by training using GPUs and
high-end DSP and FPGA kits.

20. THANKYOU