Server Start Guide - Gachon-Graduation-work/Muzzle_Detection_Project GitHub Wiki

Introduction

Server consists of YoloV3-based detector and ResNet-based classifier .

• YoloV3 is used to detect dog & muzzle.
• ResNet is used to classify whether a dog is fierce dog or not.

Dependencies

• PyTorch
• CUDA
• cuDNN
• Anaconda

Python 3.8 version with PyCharm is used for this project

Start Guide For Socket Programming

1. Install Python and Python development tool(ex. PyCharm) on your computer.
2. Downlaod project.
3. Go to main.py from Graduation_Project_Python.
4. Set your Port Number and IP Address to communicate with Client.
5. Run main.py
6. Go to MainActivity.java from Graduation_Project_App.
7. Set your Port Number and IP Address to communicate with Server.
8. Run Application from your phone.
9. Take picture then App will send picture to Server.
10. Then Server will send result picture to your App.

More details

This project offer pre-trained YoloV3 and ResNet model.
If you want to use newly-trained model, then Please refer YoloV3 part and ResNet part.