Scribd
Upload
172 views17 pages

Eco Bloom

ECO-BLOOM is an automated greenhouse solution utilizing IoT and machine learning to enhance agricultural efficiency and sustainability. The system aims to provide real-time monitoring, predictive analytics for crop health, and a user-friendly interface for control. It addresses global challenges in food production by optimizing resource usage and improving crop yields while reducing waste.

Uploaded by

krishanupal22
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
172 views17 pages

Eco Bloom

ECO-BLOOM is an automated greenhouse solution utilizing IoT and machine learning to enhance agricultural efficiency and sustainability. The system aims to provide real-time monitoring, predictive analytics for crop health, and a user-friendly interface for control. It addresses global challenges in food production by optimizing resource usage and improving crop yields while reducing waste.

Uploaded by

krishanupal22
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/ 17

ECO-BLOOM

An automated and intelligent greenhouse solution powered by IoT and ML


PRESENTED
BY
KRISHANU PAL

4th Year, ECS-501021010005


Ms. SUPARNA MAITY PRANTIK ROY
Asst. proff., Dept. of ECS 4th Year, ECS-501021010007
RAVI KUMAR

4th Year, ECS-501021010009


GUIDED SAYAN HALDER
BY
4th Year, ECS-501021010012

Friday, 22 November 2024 2


WHAT IS A GREENHOUSE?
▪ Overview:
➢ A structure designed to create a controlled environment for plant growth.
➢ Built to capture and retain sunlight, ensuring an optimal plant
temperature.
➢ Shields plants from adverse weather conditions, pests, and diseases.
➢ Allows for year-round cultivation, irrespective of outdoor climate.

▪ Key features:
➢ Transparent walls (glass or plastic) allow sunlight in.
➢ Adjustable vents for air circulation.
➢ Maintains warmth even in cold climates.
➢ Uses drip or mist systems for watering.

“Greenhouses are vital for modern agriculture, promoting


sustainability and efficiency in crop production.”
Friday, 22 November 2024 3
WHY DO WE NEED SMART GREENHOUSES?
▪ Global Challenges:
➢ Rising food demand due to population growth.
➢ Unpredictable climate changes affecting crops.
➢ Wastage of water and other resources in traditional
farming.
▪ Existing Issues:
➢ Manual monitoring leads to inefficiency.
➢ Lack of predictive insights for crop health and type.

▪ Our Aim:
➢ Build a smart greenhouse system that uses IoT for
automation and ML for predictive analytics to improve
farming outcomes.

Friday, 22 November 2024 4


WHAT WE AIM TO ACHIEVE?
Implement IoT-based real-time
monitoring and automation.

Develop ML models for:


• Predicting plant survivability.
• Recommending crops based on environmental data.

Design a user-friendly dashboard for


visualisation and control.

Friday, 22 November 2024 5


SYSTEM ARCHITECTURE
▪ IoT Layer:
➢Sensors collect real-time data: temperature, humidity,
soil moisture, light intensity, CO₂ levels.
➢Actuators control irrigation, ventilation, and lighting.
▪ Communication Layer:
➢Data is sent to a cloud platform via Thinkspeak or Blynk
IoT.
▪ Machine Learning Layer:
➢Models analyze data for predictions and
recommendations.
▪ User Interface:
➢Dashboard visualises data, predictions, and system status.
➢Remote visualisation/control features allow user
interaction.
Friday, 22 November 2024 6
KEY HARDWARE AND SOFTWARE
▪ Sensors:
➢Temperature: DHT11
➢Humidity: DHT11
➢Soil Moisture: Capacitive Soil Moisture Sensor
(HW-88)
➢Light Intensity: BH1750 or LDR
➢pH Level: Analog pH Sensor
➢CO₂ Levels: MQ-135
▪ Actuators:
➢Water pumps for irrigation.
➢Fans for ventilation.
➢LED grow lights.
▪ Controller:
➢Microcontroller: ESP8266
▪ Software Tools:
➢Data Transmission: Wi-Fi.
➢Cloud Platforms: Thinkspeak IoT, Blynk IoT.
Friday, 22 November 2024 7
➢Machine Learning: Jupyter notebook.
DATA THAT DRIVES THE MODELS
▪ TEMPERATURE:
▪ Plant Survivability Dataset:
•Tropical plants (e.g., orchids,
➢Features: Temperature, Humidity, bananas):
Soil Moisture, Light, CO₂, pH. ➢Day: 25–30°C (77–86°F)
➢Target: Survivability (Yes/No). ➢Night: 18–22°C (64–72°F)
▪ Crop Recommendation Dataset: •Temperate plants (e.g., tomatoes,
➢Features: Temperature, Humidity, cucumbers):
Light, pH. ➢Day: 20–25°C (68–77°F)
➢Target: Crop Type (e.g., Tomato, ➢Night: 15–18°C (59–64°F)
Spinach, etc.). •Cold-climate plants (e.g., lettuce,
spinach):
➢Day: 15–20°C (59–68°F)
➢Night: 10–12°C (50–54°F)

Friday, 22 November 2024 8


CONT’D…
▪ SOIL MOISTURE:
➢ Vegetables (e.g., tomatoes, peppers): Keep soil evenly moist but not waterlogged (60–
70% field capacity).
➢ Herbs (e.g., basil, parsley): Moderate soil moisture (40–60% field capacity).
➢ Cacti and succulents: Dry soil between watering (20–30% field capacity).

▪ HUMIDITY:
➢ Tropical plants: 70–90% relative humidity
➢ Temperate plants: 50–70% relative humidity
➢ Arid-region plants (e.g., cacti, succulents): 30–50% relative humidity

▪ VENTILATION:
➢ Proper ventilation ensures temperature and humidity control, preventing fungal diseases
and promoting gas exchange.

Friday, 22 November 2024 9


IMPACT ON FARMING
▪ Environmental:
➢ Reduced water and energy
waste.
➢ Improved resource allocation.

▪ Economic:
➢ Increased crop yield.
➢ Lower operational costs.

▪ Social:
➢ Enhanced food security.
➢ Scalable urban farming
solutions.
Friday, 22 November 2024 10
BARRIERS TO SUCCESS
PROBLEMS
▪ Collecting clean, consistent data to
design the predictive model.
▪ Information collection for the
presented idea.

SOLUTIONS
Snapshots of the datasets
▪ Collected the information by going
through various resources like
journals, research works, searching
for data over the internet and
Generative AI.
▪ Fabricated the dataset based on
collected information. Friday, 22 November 2024 11
RESULT ANALYSIS

Classification report for the plant Classification report for the crop
survivability model using random forest prediction model using Decision tree

Friday, 22 November 2024 12


CONT’D…
ML MODEL PLANT CROP PREDICTION
SURVIVABILITY
RANDOM FOREST 91.50 % 79.00 %
SUPPORT VECTOR 75.00 % 40.50 %
MACHINE
LOGISTIC REGRESSION 76.00 % 46.50 %
DECISION TREE 83.00 % 79.50 %
K-NEAREST NEIGHBOUR 72.50 % 44.00 %

“RANDOM FOREST and DECISION TREE performed better in both predictive models.”
Friday, 22 November 2024 13
VISION FOR THE FUTURE
▪ Incorporate AI for pest detection, disease
prediction, etc.
▪ Use drone technology for large-scale monitoring.
▪ Blockchain for secure greenhouse data storage.
▪ Extend system capabilities for different crop
types.
▪ Expand the system to optimise growth for a wider
variety of crops.
▪ Incorporate renewable energy sources like solar
panels for powering IoT devices.
▪ Scale the project to support urban and rural
farming communities.

Friday, 22 November 2024 14


SUMMING IT UP
▪ This system provides predictive insights and automation.
▪ A scalable solution for sustainable farming.
▪ This project integrates IoT and Machine Learning to create a smart greenhouse,
addressing key agricultural challenges.
▪ It automates monitoring, optimizes resource usage, and enhances crop health through
predictive insights.
▪ The system ensures sustainable farming by reducing water and energy wastage while
adapting to environmental changes.
▪ Scalable and impactful, it supports global food security and resource optimization,
paving the way for efficient agriculture.

Friday, 22 November 2024 15


LET’S DISCUSS

"We are happy to answer your questions!”


Friday, 22 November 2024 16
Friday, 22 November 2024 17

You might also like