Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 18

ANIMAL INTRUSION

DETECTION IN FARMLANDS
INTRODUCTION
 Agriculture is the main source of livelihood in different parts of the world.

 Intrusion of animals is a special challenge for the farmers throughout the world

 By incorporating this system the animal intrusion can be prevented and


agricultural process can be easily monitored.
OBJECTIVE
• Animal intrusion in farmlands is one of the greatest problems for farmers who
considers farming as their way of livelihood.
• Over the years, attempts have been made to develop systems that can identify and
give signal to the farmer and thereby make a solution for this. Obviously, such an
interface would yield great benefits.
• The main objective of this Project is to ensure effective and reliable approaches
for recognizing animals and birds which are hazardous to the act of doing
agriculture
SAMPLE DATASETS USED
The dataset used here is a collection of image data that contains various images of
animals.
Data is collected from Kaggle website : an open source data collection platform
and by scraping the data from Google Images using Selenium Web Driver
Kaggle link : https://www.kaggle.com/datasets/ashishsaxena2209/animal-image-
datasetdog-cat-and-panda
TECHNOLOGIES USED

 Front End: Python

 IDE: Visual studio code


MODULE DESCRIPTION
Animal Detection: The proposed method uses the YOLO object detection model
to ascertain presence of wild animals in images and video. The model is fine-
tuned for identifying different entities—humans, and different types of animals.
Alerting System: Alerts the user when detected an animal
Email Notification : Notify the user via email when detected an animal
SMS Notification: Notify the user via sms when detected an animal
EXISTING SYSTEM
The existing system install electrical fences to the farms and uses big
floodlights in the farm. .
 Other traditional methods followed by farmers is to hire guards to keep an
eye on crops and prevent wild animals.
LIMITATIONS OF EXISTING SYSTEM
Electric fencing is used, It is harmful to animals.
Installing an electrical fence is much costlier for huge farms and kills so many
animals, which is offensive legally and affects biodiversity
Farmer has to visit the farmland to check his crops if he came to know about
animal intrusion.
PROPOSED SYSTEM
The system will detect animals with good amount of accuracy.

Alerts the user when detected an animal

Sends the email regarding intrusion


ALGORITHMS USED

YoloV3 : This is an implementation of the object detection architecture "YOLOv3"


using the deep learning framework "Pytorch". YOLOv3 is a neural network
architecture for performing object detection on images and videos, it's a
refinement of the previous YOLO9000.
YoloV3
• The YOLOv3 algorithm first separates an image into a grid. Each grid cell
predicts some number of boundary boxes (sometimes referred to as anchor
boxes) around objects that score highly with the aforementioned predefined
classes.
• YOLO v3 uses a variant of Darknet, which originally has 53 layer network
trained on ImageNet. For the task of detection, 53 more layers are stacked
onto it, giving us a 106 layer fully convolutional underlying architecture for
YOLO v3.
• The detections are made at three layers 82nd, 94th and 106th layer.
SCREEN SHOTS
SCREEN SHOTS
SCREEN SHOTS
CONCLUSION & FUTURE SCOPE
Thus this project uses Convolutional Neural Network (CNN) algorithm to
detect wild animals.
The algorithm classifies animals efficiently with a good number of accuracy
and also the image of the detected animal is displayed for a better result so
that it can be used for other purposes such as detecting wild animals entering
into human habitat and to prevent wildlife poaching and even human animal
conflict.
This work can be further extended by sending an alert in the form of a
message when the animal is detected to the nearby forest office. We can also
add more number of species to the system for future works.Furthermore it
can be used to reduce human wildlife conflict and also animal accidents.
REFERENCES

• Xie, Z., A. Singh, J. Uang, K.S. Narayan and P.Abbeel. Multimodal blending for high-accuracy Instance
cognition. In: 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems. Tokyo: IEEE
2013, pp. 2214-2221. ISBN 978-1-4673-6356-3. DOI: 10.1109/IROS.2013.
• Tiber Trnovszky, Patrik Kamencay, Richard Orjesek, Miroslav Benco, Peter Sykora. Animal recognition
system based on convolutional neural network.
• Ahonen, T., Hadid, A., Pietikainen, and M.: Face description with local binary patterns:Application to face
recognition. IEEE TPAMI 28(12), 2037-2041 (2006).
• Burghardt, T., Calic, J.: Real-time face detection and tracking of animals. In:Neural Network Applications in
Electrical Engineering. pp. 27{32. IEEE (2006).
• Felzenszwalb, P.F., Girshick, R.B., McAllester, D., Ramanan, D.: Object detection with discriminatively
trained partbased models. IEEE TPAMI 32(9), 1627-1645(2010).
• He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR. pp. 770-778
(2016).
Thank you

You might also like