Securade.ai HUB is a generative AI based edge platform for computer vision that connects to existing CCTV cameras and makes them smart. It uses natural language text and generative AI to automatically train and fine-tune state-of-the-art computer vision models on the edge. This eliminates costly data labelling and annotations work typically required in training new models. Thus, enabling us to deploy a custom accurate model per camera feed.
🤖 Zero-Shot Learning - No manual data labeling needed - train models using natural language descriptions
🎯 Real-Time Detection - Live monitoring of CCTV feeds for safety and security incidents
👷 PPE Detection - Automated detection of hardhat, vest, and mask compliance
⚡ Proximity Alerts - Monitor safe distances between workers and machinery/vehicles
🚫 Exclusion Zones - Configurable restricted area monitoring and access control
📊 Analytics Dashboard - Real-time safety metrics and violation tracking interface
🎥 Multi-Camera Support - Compatible with D-link, Tapo, TP-Link, Axis and HikVision cameras
🔔 Instant Notifications - Real-time Telegram alerts for detected violations
🎭 Privacy Protection - Automatic face masking to maintain worker privacy
💻 Edge Processing - Local processing without cloud dependency
git clone https://github.com/securade/hub.git
cd hub
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txtYou can check if the installation was successful by running:
python securade.py --help
usage: securade.py [-h] [--config CONFIG] [--cpu] [--no_activity_alert NO_ACTIVITY_ALERT] [--server] [--version]
options:
-h, --help show this help message and exit
--config CONFIG Load config from file
--cpu Use the OpenVINO Runtime for inference on CPU
--no_activity_alert NO_ACTIVITY_ALERT
Time in seconds after which a no activity alert is raised
--server Run the Securade web server application
--version show program's version number and exitOnce the HUB is installed you will need to configure the streamlit web server.
mkdir .streamlit
cp config.toml .streamlit
cp secrets.toml .streamlitYou can then run the web server to configure the CCTV cameras and policies:
python securade.py --server
--------------------------------------------------------------------------
# ##### #
# # # ###### #### # # ##### ## ##### ###### # # #
# # # # # # # # # # # # # # # # #
# ##### ##### # # # # # # # # # ##### # # #
# # # # # # ##### ###### # # # ### ####### #
# # # # # # # # # # # # # # # ### # # #
# ##### ###### #### #### # # # # ##### ###### ### # # #
--------------------------------------------------------------------------
Press Enter to exit ...
You can now view your Streamlit app in your browser.
Network URL: http://192.168.10.147:8080
External URL: http://58.182.134.244:8080The default password is pass, you can change it in the secrets.toml file that you have in the .streamlit folder.
To use Telegram for notifications, you will need to get an api_id and the chat_id for the channel or group you want to send the notifications.
You can get the api_id by creating a new application here. The chat_id can be obtained
by following the instructions here. Make sure you add your bot to the channel or group and set group admin rights for it.
You can watch a detailed demo and turorial on how to use the HUB here.
Installing the HUB on an edge device with GPU is significantly more involved as we need to ensure that correct versions of Jetpack, CUDA, cuDNN, Torch and Torchvision are installed. You can read the detailed instructions on the wiki.
The HUB should work with any Jetson device with atleast 8 GB of memory. It has been tested to work on Lenovo ThinkEdge SE70 Edge Client and the NVIDIA Jetson AGX Orin Developer Kit.
The HUB includes a flexible plugin system that allows you to extend its image processing capabilities. You can create custom plugins to add new computer vision features, image processing algorithms, or integrate with other AI models.
The HUB comes with several built-in plugins:
- YOLO Object Detector: Default object detection using YOLOv7
- Edge Detector: Basic edge detection using OpenCV
To create a new plugin, add a Python file to the plugins directory that implements the BasePlugin interface:
from plugins.base_plugin import BasePlugin
class MyPlugin(BasePlugin):
SLUG = "my_plugin"
NAME = "My Custom Plugin"
def run(self, image, **kwargs):
# Your image processing code here
return processed_imageThe plugin will automatically be discovered and appear in the HUB interface's plugin selection dropdown.
For detailed documentation on creating plugins, including examples and best practices, see the plugins/README.md.
Securade.ai HUB is open-source and available under the GNU AGPL license. You can freely use it on your own edge devices or servers. If you are looking to bundle it and distribute to others, you will need a commercial license.
You can get the Securade.ai Commercial License for a nominal fees of 99 SGD per month. Securade.ai is a tech4good venture and the commercial license allows you to deploy the HUB on an unlimited number of devices and servers.
|
|
|
|
|
|
|
|
|