Poster Template

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 1

Artificial Intelligence and IoT in Agriculture

Gundu Hemanthsai, 4th Year AI, Vidya Jyothi Institute of Technology


Introduction Methods Methods Cont. Results

According to a UN survey, the 1. Cotton Leaf Disease Detection


General Method Convolutional Neural Network
world's population will rise from 7.5 (ANN)
billion to 9.7 billion by 2050.
In order to feed the entire
population the food production
2. Crop Recommendation
must be increased by 60%.
This means that farmers must
accomplish more with less. Existing
agricultural methodologies are not
precise enough to perform tasks like 1. Crop Disease Detection (Eg:
detecting crop diseases and Cotton Leaf Disease Detection)
CNN Architecture: IoT Use Cases
choosing the best crop among Four (2D Convolution Layers +
different crops, and deal with rising MaxPooling Layer). Sensors determine soil fertility
in real time, ML algorithms
demand. Followed by ANN with 2 layers suggest to use fertilizer or not.
Output Layer: 1 Neuron Real time temperature,
The solution for above problem is (Sigmoid Activation, output='0' moisture, & humidity
using AI and IoT in Agriculture to for diseased & '1' for fresh leaf) monitoring
improve the quality and efficiency
of the production through improved 2. Crop Recommendation Conclusions
decision making processes. Artificial Neural Network (ANN) To predict the best suitable crop
based on the following features: CNN Model performs best for
NPK Ratio
Objectives Temperature
Cotton Leaf Disease Detection.
SVM Model performs best for
Humidity
Crop Recommendation.
To develop Machine Learning & PH & Rainfall
Deep Learning models for ANN Architecture:
detection of Crop Diseases and I/P Layer: 150x150X3 Neurons Algorithms: References
Crop Recommendation. (Input Image 150x105 px). Support Vector Machine (SVM
Identify the most efficient
Hidden Layers: 3 (ReLU Classifier) https://fortune.com/2020/10/05
Activation). Logistic Regression
supervised learning algorithm. /a-i-precision-agriculture-deere/
O/P Layer: 1 Neuron (Sigmoid Random Forest Classifier
Identify the use cases where IoT https://sciencedirect.com/scienc
Activation, Output='0' for
can be integrated. diseased & '1' for fresh leaf) e/article/pii/S258972172030012X

You might also like