AI Based Farming Assistant

Under the guidance of BY BATCH-15

P. Pavan kumar P.USHA SREE

M.Tech, M.B.A SARANYA SAI.K

SR.ASSISTANT PROFESSOR Y.MANOJ KUMAR

K.JNANESWARI
Contents
o Abstract

o Introduction

o Problem statement

o Literature Survey

o Need for Project

o Work to be done

o Advantages and Disadvantages

o Applications
Abstract
o This paper presents an AI-driven farming assistant system designed to enhance agricultural efficiency
and productivity. The system utilizes a Raspberry Pi as its central processing unit, equipped with
sensors to measure critical environmental parameters, such as humidity, temperature, and soil
moisture. By leveraging a Convolutional Neural Network (CNN) model, the assistant analyzes data
to provide accurate predictions and actionable insights for crop management. The CNN is trained to
recognize patterns in environmental factors and can predict crop health or disease likelihood,
irrigation needs, and optimal growing conditions. This low-cost, automated system aids farmers in
making data-driven decisions, improving yield and resource management while minimizing labor
and water usage. The proposed solution aims to be a scalable and accessible tool for sustainable
farming practices, particularly in small to medium-sized agricultural setups.

.
Introduction
o To enhance crop yield with minimal initial information, you can develop a smart farming
solution that assists farmers by providing actionable insights on crop conditions, pesticide
usage, and overall crop maintenance.

o The project can begin by collecting data through affordable IoT sensors to monitor key
parameters such as soil moisture, temperature, and pH levels, and by utilizing satellite imagery
for crop health monitoring. This data, combined with inputs from farmers, can be analyzed to
generate basic recommendations.

o For example, a rule-based system could advise on optimal pesticide quantities based on crop
type and environmental conditions. A simple user interface, like a mobile app, would enable
farmers to receive these recommendations and alerts in real-time, with support for local
languages to ensure accessibility.
Problem Statement
o Traditional farming relies heavily on manual monitoring of environmental factors like soil moisture,
temperature, and humidity, which is labor-intensive, time-consuming, and often inaccurate,
particularly for smaller farms. Additionally, unpredictable weather and climate change complicate
farmers' ability to make informed decisions, leading to inefficient resource use and suboptimal crop
management. This project proposes an AI-based farming assistant using a Raspberry Pi and a
Convolutional Neural Network (CNN) to automate data collection and analysis of critical
environmental conditions. The system provides real-time insights and recommendations to help
farmers optimize irrigation, predict crop health, and enhance resource management, offering an
affordable solution for precision agriculture and sustainable farming practices.
LITERATURE SURVEY
PAPER TITLE YEAR OF PUBLICATION DESCRIPTION

Revolutionizing Farming: GAN- 04 April, 2024 An advanced farming system using

Enhanced Imaging, CNN Disease GAN imaging, CNN disease
Detection, and LLM Farmer detection, and LLM assistance to
enhance crop management and
Assistant
support farmers.
AI-CABB: An AI-Enabled 25 March, 2024. AI-CABB uses machine learning to
Integrated Pest Detection System detect pests in cabbage, enhancing
for Enhancing Cabbage cultivation efficiency and yield
through precise, automated
Cultivation Using Machine
monitoring.
Learning
Enhancing crop health using deep 21 May, 2024 Deep learning techniques analyze
learning techniques plant data to detect diseases early and
optimize practices, improving crop
health.
Drone-Based Weed And Disease 06 March, 2024. A drone-based system using YOLOv8
Detection In Agricultural Fields for efficient weed and disease
To Maximize Crop Health Using detection, maximizing crop health in
agricultural fields.
a Yolov8 Approach
PAPER TITLE YEAR OF PUBLICATION DESCRIPTION
Advancing Precision Agriculture: 11 June, 2024. Advancing Precision Agriculture
AI, IoT, and Machine Learning involves using AI, IoT, and machine
Solutions for Optimal Crop Yield learning to optimize crop yield through
data-driven decision-making and
automated farming practices.

Deep learning based identification 05 August 2023 A deep learning-based system for
of solid waste management in smart cities identifies and separates
smart cities through garbage solid waste, improving garbage
management and monitoring for
separation and monitoring
efficient waste disposal.

Water pollution accident control 08 September 2011 This refers to strategies to prevent and
and urban safety water supply manage water pollution, ensuring safe
explain the meaning and reliable urban water supplies.

An AI approach for herbal plant 02 July 2024 An AI-driven system for detecting and
detection and recognition recognizing herbal plants from images,
enabling automated identification .

Landslide detection system using 20 February 2020 A Raspberry Pi-based Landslide

computer vision approach and Detection System uses computer vision
Raspberry Pi to monitor terrain changes and send
real-time alerts, helping prevent
disasters.
 Need for project
o This AI-based farming assistant project leverages the capabilities of a Raspberry Pi to
automate and optimize plant care by monitoring key environmental factors such as
temperature, humidity, soil moisture, and plant identification. Using sensors like the
DHT22 for temperature and humidity, a moisture sensor for soil hydration, and a camera
module for plant classification via a Convolutional Neural Network (CNN), the system
provides real-time data on plant conditions. By identifying plant species and continuously
monitoring the environment, this solution can help farmers make informed decisions,
improve crop health, conserve resources, and ultimately enhance agricultural
productivity.
Work to be done
o The AI-based farming assistant project involves assembling hardware components like a Raspberry Pi,
camera module, DHT22 temperature and humidity sensor, and moisture sensor to monitor
environmental factors such as soil moisture, temperature, and humidity, while using image recognition
to classify plant types through a Convolutional Neural Network (CNN). Tasks include integrating the
sensors, capturing data, training or fine-tuning the CNN model for plant classification, and developing
software to collect and display real-time sensor data. Additionally, the system should be optimized for
performance on the Raspberry Pi, with potential features like cloud integration, notifications, and a user-
friendly web interface for remote monitoring.

o We began by setting up a Raspberry Pi and interfacing it with a computer. After this, we installed the
necessary software packages.
o And commands to prepare the Raspberry Pi for further tasks. Next, we
connected a camera module to the Raspberry Pi and ensured all required
dependencies were installed to facilitate image capturing.

o Once the setup was complete, we gathered a diverse set of datasets,

specifically related to crop identification. Using these datasets, we trained
a machine learning model to process and analyze the images effectively.

o To validate the model's performance, we conducted rigorous testing by

inputting images of various crops and evaluating the model's ability to
classify them accurately. This approach helped us ensure that the system
is capable of real-time crop detection, marking the culmination of our
project work.
Cont...
COMPONENTS

o Raspberry pi

o Camera module

o Moisture sensor

o Temperature sensor

o Humidity sensor
 CROPS DISEASES

TOMATO 1. Tomato_Early_Blight
1.
Symptoms:
Brown spots with concentric rings on leaves, stems, and fruits. Severe infections can cause leaf drop
and reduced yield.

DEFORMATION CAUSES:
 Fungal Infection - Alternaria solani
 Environmental Conditions - Warm and humid conditions (20–30°C or 68–86°F) and poor air
circulation
 Infected Plant Debris.
 Nutrient Deficiency - Potassium and nitrogen deficiencies
2.
2. Tomato Late_blight
Symptoms:
Yellowing and curling of leaves, stunted plant growth, and reduced fruit production. Spread by
whiteflies.

DEFORMATION CAUSES:
 Pathogen Presence - Phytophthora infestans.
 Environmental Conditions - High humidity, cool temperatures, prolonged leaf wetness
 Spore Dissemination - Wind and rain, human activity
 Poor Plant Health Management
 CROPS DISEASES

3.Tomato Leaf_Mold

Symptoms:
 Olive-green to yellowish spots on upper leaves.
 White/gray fungal growth on the underside of leaves.
3
DEFORMATION CAUSES:
.  Overhead Watering
 Infected Plant Material
 Lack of Crop Rotation

4. Tomato septoria leaf_spot

Symptoms:
 Small, circular lesions with dark brown borders and light gray centers on leaves.
 Yellow halos around the spots.

DEFORMATION CAUSES:
4  Pathogen Presence - Septoria lycopersici.
 Environmental Conditions – Moisture, humidity.
 Poor Air Circulation
 Inadequate Sanitation
 CROPS DISEASES

5. Tomato___Spider_mites Two-spotted_spider_mite
5
Symptoms:
 Tiny yellow or white spots on leaves.
 Silken webbing on the undersides of leaves and stems.

DEFORMATION CAUSES:
 Environmental Conditions - Hot and dry weather , low humidity.
 Plant Stress - Drought stress, nutrient deficiency
 Overuse of Insecticides.
 Lack of Natural Predators.
6.
6. Tomato___Target_Spot

Symptoms:
 Concentric, circular lesions with dark brown centers and yellow halos on leaves.
 Dark spots on stems and fruits.

REMEDIES:
 Pathogen Presence - Corynespora cassiicola.
 Crop Management Practices - Overhead Irrigation.
 Prolonged Leaf Wetness.
 Susceptibility of Tomato Varieties.
 CROPS DISEASES

7. Tomato_Yellow_Leaf_Curl_Virus

Symptoms:
 Yellowing and curling of leaves (upward and inward).
7.  Stunted growth and reduced plant size.

DEFORMATION CAUSES:
 Causative Agent - virus belonging to the genus Begomovirus
 Vector Transmission – Whiteflies.
 Poor Crop Management - Infected plant material , lack of sanitation, ineffective pest control.

8. Tomato_mosaic_virus

8. Symptoms:
 Curling, deforming leaves and stunted growth.
 Reduced fruit size with discoloration and poor quality.

DEFORMATION CAUSES:
 Virus Transmission - Seed-Borne Transmission, Soil and Debris.
 Human Activities – lack of crop rotation.
 Environmental Factors.
 Host Plant Susceptibility.
 CROP DISEASES

CRON 1. Corn_(maize)___Cercospora_leaf_spot Gray_leaf_spot

1. Symptoms:
 Gray or brown lesions with yellow halos on leaves.
 Long, narrow lesions that may turn gray with dark borders.

DEFORMATION CAUSES:
 Pathogen - Cercospora fungus.
 Environmental Conditions - Warm, humid weather favors fungal growth.
 Infected Plant Debris - The fungus survives in soil and plant debris, leading to recurrent
infections.

2.
2 Corn_(maize)___Common_rust

Symptoms:
 Small, circular, orange to red pustules on leaves.

DEFORMATION CAUSES:
 Pathogen Presence - Puccinia sorghi.
 Environmental Conditions - Cool temperatures, High humidity.
 Wind-Borne Spores
 CROP DISEASES

3. Corn_(maize)___Northern_Leaf_Blight

3. Symptoms:
 Long, grayish-green lesions on leaves with dark brown to gray centers.
 V-shaped lesions at leaf tips.

DEFORMATION CAUSES:
 Pathogen Presence - Setosphaeria turcica.
 Environmental Conditions – Temperature, Rainfall.
 Nutrient Deficiency and Stress.

4.
4.Corn_(maize)___healthy

Symptoms:
 Strong, green leaves with no lesions or discoloration.

DEFORMATION CAUSES:
 Soil Fertility - Soil pH.
 Watering and Irrigation - Adequate Moisture, Well-Drained Soil.
 Crop Rotation
 CROP DISEASES

1. Anthracnose (Colletotrichum spp.)

CHILLI
Symptoms:
1.  Dark, sunken lesions on fruits, stems, and leaves.
 In severe cases, lesions may coalesce, leading to fruit rot.

DEFORMATION CAUSES:
 Fungal Pathogen - Colletotrichum fungus.
 Infected Plant Debris.
 Improper Spacing.
 Inadequate Crop Rotation

2. 2. Chili Leaf Curl Virus (ChiLCV)

Symptoms:
 Upward curling and distortion of leaves.
 Stunted plant growth and reduced fruit set.

DEFORMATION CAUSES:
 Vector Transmission - The primary cause of ChiLCV spread is the Bemisia tabaci, a type of
whitefly.
 Movement of Infected Plant Parts.
 Lack of Disease Management Practices.
Pest And Pesticides

Commonly seen pest and pesticides:

1. Tomato

Pest - Fruit borer

pesticide – Pyrethrins

2. Corn

Pest - Fall army worm

Pesticide - Emamectin benzoate

3.Chilli

Pest – Aphids

Pesticide - Imidacloprid
Advantages
o Increased Efficiency.

o Better Crop Managementns.

o Early Disease Detection.

o Resource Conservation5.

o Scalability:.

o Data-Driven Insights.

o Remote Monitoring.

o Cost-Effective.
Disadvantages
o Initial Setup Cost.

o Maintenance and Calibration.

o Model Accuracy.

o Limited Processing Power.

o Dependency on Internet/Power.
Applications
o Precision Agriculture.

o Crop Health Monitoring.

o Smart Irrigation Systems.

o Farm Automation.

o Yield Prediction and Crop Planning.

o Climate Monitoring.

o Remote Farm Monitoring.

o Sustainable Farming.
CODE...
References
o https://agriculture.vic.gov.au/biosecurity/plant-diseases/vegetable-diseases/tomato-yell
ow-leaf-curl-virus

o https://www.kaggle.com/datasets/sashisekeran/tomato-and-chilli-dataset-with-annotatio
ns

o https://www.kaggle.com/datasets/abdallahalidev/plantvillage-dataset
Thank You