/SRS-Project

Safety, Rescue & Security Object detection Models

Primary LanguageJupyter Notebook

Safety, Rescue & Security (SRS) Project

Object detection is a computer technology related to computer vision and image processing that deals with detecting instances of semantic objects of a certain class (such as humans, buildings, or cars) in digital images and videos. With the development of deep learning, object detection has gained lots of improvements. The SRS project is a project which shows the use cases of object detection in the environments regarding safety, rescue operations and security applications. The SRS project breaks down into three (3) industries

  1. Safety

  2. Rescue

  3. Security

Safety

image

This section of the project aims to develop a safety object detection model for the industrial environment. The model is designed to detect potential safety hazards, such as workers without proper personal protective equipment (PPE), in real-time video streams from industrial sites.

The model uses state-of-the-art object detection techniques, YOLOV5 to accurately detect and classify objects in the industrial environment. The model has been trained on a dataset of images and videos collected from various industrial sites, and has been fine-tuned to improve performance on specific safety hazards.

The goal of this project is to improve safety in the industrial environment by providing real-time feedback to workers and supervisors on potential safety hazards. This model can be integrated into existing surveillance systems to provide an additional layer of safety monitoring.

Rescue

image

This project aims to develop a rescue object detection model for emergency situations. The model is designed to detect and locate individuals in need of rescue in real-time video streams from drones or surveillance cameras. The model can be used by first responders and rescue teams to quickly identify and reach victims in a timely manner.

The model uses state-of-the-art object detection techniques, YOLOv5 to accurately detect and classify objects in images and videos captured during emergency situations. The model has been trained on a diverse dataset of images and videos collected from people in water bodies carrying out recreational activities to econoomic activities.

The goal of this project is to improve the effectiveness and efficiency of rescue operations by providing real-time information to first responders on the location of victims. The model can also be used to monitor the progress of rescue operations and make adjustments as needed.

Security

Surveillance Camera Records Deadly State Street Shootout (73)

This project aims to develop a security object detection system that uses computer vision algorithms to detect and classify objects in real-time video streams.

The goal of the project is to automatically identify and track objects of interest, such as people and guns, in order to detect and alert security personnel of potential security breaches.

We used a deep learning-based object detection algorithm, such as Faster RCNN, to train a model on a dataset of labeled security footage. The model is then deployed to a network of CCTV cameras or surveillance systems for real-time object detection with additional features such as object tracking.

Media Folder

The media folder conatins some testing images and our final model result images.