It mainly solves the two major challenges faced by surveillance systems in public places: privacy protection and intelligent image recognition. In terms of protecting personal privacy under these challenges, privacy encryption technology is proposed to use various encryption algorithms and steganography technology for face information encryption. Decryption can ensure that the confidentiality and concealment of face information can be maintained when presenting in the image. In terms of intelligent image recognition, the deep learning model is used to separate the human body from the image and recognize its movements, which can instantly identify whether it is in danger. In the middle, emergency incidents can be dealt with. This research aims to balance the public security needs and the protection of personal privacy and can be applied to various scenarios.
fall_encrypt.mp4
- Clone project
git clone https://github.com/ytl0623/yolov7-GDPR.git
- Create a new conda environment
conda create -n yolov7 python==3.9
- Activate conda environment
conda activate yolov7
- Install package
pip install -r list.txt
- Install Detectron2
cd detectron2
python setup.py install
- Run any of the Python scripts you would like to try out
python encrypt_image/video decrypt_image/video