Copyright 2
Copyright 2
MEMBERS
Palak Baisware
Sadiya Tahsin Ansari
Manasvi Borkar
Bharat Taklikar
Sahil Bansod
This system is engineered under the domain of the Internet of Things (IoT) along with the help of
technology of Machine learning covering the concept of deep learning, to detect the presence of
animals and humans in a specific area, minimizing the chances of unexpected encounters. By
providing timely alerts, it allows individuals to take necessary precautions, thereby preventing
accidents before they occur. Our device is a society-based innovation that focuses on safeguarding the
entities, from any harm which can be caused by a human to an animal and from an animal to a human.
The available device which has been under the same domain working in this sector shares an accuracy
rate of Approximately 82% with specific drawbacks. Those Drawbacks consider the range of the
product for detection, accuracy, speed, and efficiency. The systems that are used in the Smart cars and
self-driving vehicles are effective, however as system decisions are taken without any human
supervision leading to self-accident mode of the car, unnecessary interference while driving and many
more issues. We aim to cover all those points and factors of drawback in the development of this
device.
ADA device works on our specifically designed model, following the R-CNN algorithm for sending
messages and alerts to the user. This device overcomes the factor of night hindrance as in some
specific conditions night vision is important to prevent accidents which is majorly contributed by the
lack of street lights in these specific regions. An animal detection and alerting device primarily
focuses on monitoring animal activity and reducing human-animal collisions. By continuously
tracking the movement of animals in real-time and the frequency of crossing the road. This data helps
wildlife managers and researchers monitor and manage animal populations, providing information
that can guide conservation efforts or inform decisions about habitat protection. Additionally, the
system can detect abnormal behavior, like animals in distress, alerting authorities to potential health or
safety issues.
In terms of reducing human-animal collisions, these devices play a crucial role in enhancing road
safety. By detecting animals near or on roadways, the system can provide early warnings to drivers,
allowing them to slow down or take precautions, significantly lowering the risk of accidents. This is
especially important in high-risk zones, such as rural roads or wildlife corridors, where animals
frequently cross. Our aim is to ensure the prevention of any human- animal collisions, safeguarding
both the entities and providing a safe driving environment.
IDEA:- ANIMAL DETECTION AND ALERTING DEVICE (ADA DEVICE)
Here, we are going to use Raspberry Pi and the specifically designed software model while following the
R-CNN algorithm and Various factors in our software which helps in the working of this device. For the
issues raised by the night hindrance and lack of light facilities, to overcome them the implementation of
thermal and infrared cameras will be done.
The implementation of the device is precisely for the forest, jungle safaris, zoological garden locality,
and under-developed region which surrounds the forest region. This device is suitable for two-wheeler,
four-wheeler, three-wheeler, trucks, tempos, rains etc. to prevent any accident from happening. Below
Fig(a) shows the working architecture of our ADA Device.
********************