Scribd
Upload
6 views1 page

Assignment Format

tcs

Uploaded by

VIKRAM S
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
6 views1 page

Assignment Format

tcs

Uploaded by

VIKRAM S
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

KNOWLEDGE INSTITUTE OF TECHNOLOGY

(An Autonomous Institution)


Approved by AICTE, Affiliated to Anna University,
Kakapalayam (PO), Salem – 637 504

Department of Electronics and Communication Engineering


Sub code/Name:ET3491/EMBEDDED SYSTEM AND IOT SYSTEM
Year-III ECE-B Date:08.05.2024
PEST DETECTION AND MANAGEMENT

ABSTRACT: CODING:
import cv2
import numpy as np
This project leverages the Raspberry Pi platform to develop a pest import tensorflow as tf
detection and management system aimed at improving agricultural from tensorflow.keras.models import load_model
productivity. The system utilizes a combination of sensors, import RPi.GPIO as GPIO
from time import sleep
cameras, and machine learning algorithms to detect and identify def capture_image():
common agricultural pests in real-time. By automating the process cap = cv2.VideoCapture(0)
of pest detection and management, this solution offers a cost- ret, frame = cap.read()
if ret:
effective, efficient, and scalable approach to pest control, reducing cv2.imwrite('/home/pi/pest_images/captured_image.jpg', frame)
the reliance on manual inspections and chemical pesticides. cap.release()
return frame
model = load_model('/home/pi/pest_detection_model.h5')
def preprocess_image(image_path):
image = cv2.imread(image_path)
INTRODUCTION: image = cv2.resize(image, (224, 224))
image = image.astype('float32') / 255.0
image = np.expand_dims(image, axis=0)
Agriculture is crucial for sustaining the global population, but
return image
pests pose significant challenges, causing crop damage and def predict_pest(image_path):
processed_image = preprocess_image(image_path)
economic losses. Traditional pest management often involves prediction = model.predict(processed_image)
labour-intensive monitoring and heavy pesticide use, which can return np.argmax(prediction, axis=1)
GPIO.setmode(GPIO.BCM)
be environmentally harmful. GPIO.setup(18, GPIO.OUT)
def activate_sprinkler(duration):
GPIO.output(18, GPIO.HIGH)
This project leverages the Raspberry Pi platform to develop an sleep(duration)
automated pest detection and management system. Using GPIO.output(18, GPIO.LOW)
def main():
sensors, a camera, and a pre-trained machine learning model, image_path = '/home/pi/pest_images/captured_image.jpg'
capture_image()
the system identifies and manages pests in real-time. This pest_type = predict_pest(image_path)
approach enhances pest control efficiency, reduces manual labor, if pest_type == 0:
print("No pest detected.")
and minimizes pesticide dependence, promoting sustainable elif pest_type == 1:
print("Aphid detected. Activating sprinkler for 5 seconds.")
agriculture. By integrating affordable hardware with advanced
activate_sprinkler(5)
software, the project offers a modern solution to improve elif pest_type == 2:
print("Caterpillar detected. Activating sprinkler for 10 seconds.")
agricultural productivity and sustainability. activate_sprinkler(10)
else:
The pest detection and management system using Raspberry Pi print("Unknown pest detected.")

operates through four key steps: RESULT:


if __name__ == "__main__":
main()
1. Image Capture: The Raspberry Pi camera captures images The pest detection and management system demonstrated effective identification of pests
such as aphids and caterpillars in test environments. The use of a pre-trained deep learning
of crops at regular intervals. model allowed for accurate classification of pests, leading to timely and appropriate
management actions. This automation significantly reduced manual labor and reliance on
2. Image Processing and Pest Detection: These images are chemical pesticides, offering a sustainable approach to pest management in agriculture. The
processed and analysed using a pre-trained convolutional Raspberry Pi-based system proved to be a cost-effective and scalable solution for real-time
pest control, enhancing crop health and productivity.
neural network (CNN) to identify the type and presence of
pests.
SIMULATION OUTPUT :
3. Management Actions: Depending on the pest identified, the
system triggers appropriate actions, such as activating a
sprinkler to apply pesticides or repellents.

4. Automation and Feedback: The entire process is automated


for real-time monitoring and intervention, with the system
logging detections and actions for ongoing improvement.

STUDENTS NAME: VIKRAM S(TL),GOKULRAJ K,GOKULRAJ S,FAIMUDEEN M,NAVANEEDHAKRISHNAN V,


SHESHAIYA B

You might also like