Scribd
Upload
38 views16 pages

Flower Classification

Uploaded by

praveenlakshita31
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
38 views16 pages

Flower Classification

Uploaded by

praveenlakshita31
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 16

FLOWER

CLASSIFICATION
USING CNN
FLOWER CLASSIFICATION
USING CNN

BY
RAJESH R
[613521104032] PRAVEENKUMAR A
[613521104036] MANIGANDAN
C [613521104035]
Agenda
 Objectives
 System Requirements
 System Design
 System implementation
 Result Analysis
 Conclusion
OBJECTIVE
The objective of flower classification using
Convolutional Neural Networks (CNNs) is to
automatically categorize images of flowers into
different classes or species based on their
visual features
SYSTEM
REQUIREMENTS
1.FUNCTIONAL REQUIREMENTS
• Image Handling: The system should be able to accept and process images
of flowers in different sizes and formats.
• Classification Model: It needs to have a CNN model.
• Training and Testing: There should be a way to teach the system with
examples of labeled flower images and check how well it's learning.
• User Interface: Users should be able to easily upload flower images and
get results showing what type of flower each image is.
• Performance Metrics: The system should measure how well it's doing at
identifying flowers, using things like accuracy and precision.
• Deployment: Once ready, the system should work in a real-world setting
where it can classify images quickly and accurately.
2.NON-FUNCTIONAL REQUIREMENTS

• Security: The system should ensure data protection and secure access,
including encrypted data storage and secure communication protocols.
• Usability: The system should have an intuitive and user-friendly interface,
making it easy for users to upload images, view results, and navigate through
the application.
• Reliability: The system should consistently perform classification tasks
correctly and handle high volumes of requests without failure.
• Portability: The system should be portable across different operating systems
and environments, ensuring it can be deployed on various platforms without
significant changes.
• Efficiency: The system should utilize computational resources efficiently,
minimizing CPU and memory usage while maintaining high performance.
SYSTEM DESIGN
System design involves planning and defining the architecture, components, and
modules that make up the system. For a flower classification project using a
Convolutional Neural Network (CNN), the system design will cover data flow,
model architecture, and the interactions between various components.
OVERVIEW
The system is designed to classify images of flowers into different categories
using a CNN model. The process includes the following stages:
1.Data Collection and Preprocessing
2. Model Architecture
3. Training and Validation
4. Evaluation and Prediction
SYSTEM DESIGN
DATA COLLECTION AND PREPROCESSING:
- Data Collection: Collect flower images from a dataset (e.g., Flower Dataset with 102
categories).
- Preprocessing: Resize images, normalize pixel values, and apply data augmentation
techniques like rotation, flipping, etc.
MODEL ARCHITECTURE:
- Convolutional Layers: Extract features from input images using convolution operations.
- Pooling Layers: Reduce the spatial dimensions and maintain significant features.
- Flattening: Convert the 2D feature maps into a 1D vector.
- Dense Layers: Fully connected layers for classification based on the extracted features.
- Output Layer: Uses softmax activation to output probabilities for each flower category.
MODEL ARCHITECTURE:
SYSTEM DESIGN
Training and Validation:
- Training: Use labeled data to train the CNN model by adjusting weights
through backpropagation.
- Validation: Evaluate model performance on a validation set to fine-tune
hyperparameters and prevent overfitting.

Evaluation and Prediction:


- Evaluation: Assess model performance using metrics like accuracy, precision,
recall, and confusion matrix.
- Prediction: Apply the trained model to new images to classify them into
flower categories
ACTIVITY
DIAGRAM
USECASE
DIAGRAM
SYSTEM
IMPLEMENTATI
ON
IMPLEMENTATION MODULES
Data Collection Module Model Training Module
This module handles the collection This module handles the training
and organization of the dataset. of the CNN model.
Data Preprocessing Module Model Evaluation Module
This module includes steps for This module evaluates the model's
data augmentation and performance on the test dataset
preparation.
Model Design Module
This module defines the
architecture of the CNN model.
RESULT ANALYSIS

 The flower is predicted based on trained model


and displayed in confusion matrix.

This confusion matrix shows that predicted


flower was dandelion.
CONCLUSION
Using CNNs for flower classification worked well. The

model learned to recognize different types of flowers

accurately. It performed consistently across training and

validation, showing it can handle various flower types

effectively. To make it even better, we can adjust some

settings and add more flower pictures. Overall, CNNs are a

great way to automatically classify flowers based on their

pictures.

You might also like