Flower Recog System

A MINI PROJECT PRESENTATION
ON
FLOWER RECOGNITION SYSTEM
UNDER THE GUIDENCE OF
Dr. JAYANT NANDWALKAR
DATTA MEGHE COLLEGE OF ENGINEERING

Airoli,Navi Mumbai
Eshaan Sonavane (46)
Manas Thakare (49) Department of Computer Engineering
Dhammabhushan
Waghmare (56) University of Mumbai
Ashwin Yadav (58)
(AY 2023-24)
INDEX
 INTRODUCTION 3
 LITERATURE SURVEY 4
2.1 Literature Survey 4
2.2 Aim and Objection 5
2.3 Requirement Analysis 5
 PROPOSED SYSTEM 6
3.1 Introduction 6
3.2 Architecture/Flowchart 6
3.3 Data Set 7
3.4 Framework and Tool 8
 ALGORITHMS 9
 CONCLUSION 11
INTRODUCTION
 There are many species of flowers in the world. Some species
have many colors, such as roses. It is difficult to remember all the
names of flowers and their information. Additionally, someone
may be confused with similar flower species.
 A flower recognition system is an advanced computer vision and
machine learning application designed to automatically identify
and categorize various species of flowers from images or live
video streams.
 By analyzing the unique shapes, colors, and textures of petals,
leaves, and stems, this system provides valuable assistance to
botanists, and nature enthusiasts in identifying and cataloging
flowers in the wild or cultivated settings.
LITERATURE SURVEY
AIMS AND OBJECTIVES
 1. Develop a highly accurate and user-friendly flower recognition system capable of identifying a diverse
range of flower species based on the visual characteristics .
 2. Create an educational tool that aids in teaching about plant species and botany and contributes to
environmental monitoring and conservation efforts.
REQUIREMENT AND ANALYSIS

The requirements for the development of this project are as follows:
1)Laptop with intel iris x Graphic Card.
2)We have acquired the Dataset of the symptoms of various Flowers by
small shopkeepers and Kaggle website.
PROPOSED SYSTEM:
 We have collected a diverse dataset of flower
images that includes various species and
encompasses different lighting conditions and
backgrounds.
 Process and clean the dataset by standardizing
image dimensions, adjusting lighting variations,
and enhancing image quality.
 Convolutional Neural Network (CNN): Implement
a deep learning model, such as a CNN, known for
its effectiveness in image recognition tasks.
 Train the recognition model using the training
dataset, fine-tuning the model to optimize
accuracy.
 Continuously validate the model's performance
using the validation dataset, adjusting the model as
needed.
DATA SET
● "https://storage.googleapis.com/download.tensorflow.org/example_images/flower_photos.tg"
FRAMEWORK AND TOOLS
• Excel sheet for Flower data

• Python as core programming language
• xlsx writer and openpyxl to read and write data in python
• pandas for better visualization of data
• Html to create webtemplate
• CSS to style the website
• Javascript for backend data
• Json to store data
• GitHub to store record
• Postit for web deployment.
ALGORITHMS
 Certainly, for flower recognition, you can use the following types of algorithms:
 1. Convolutional Neural Networks (CNNs): CNNs are the most popular and effective deep learning
algorithms for image recognition, including flower recognition. They can automatically learn
features from images and classify them.
 2. Deep Learning Transfer Learning: You can use pre-trained CNN models (e.g., VGG, ResNet,
Inception) and fine-tune them for flower recognition. This approach often yields excellent results
with less training data.
 3. SIFT (Scale-Invariant Feature Transform): SIFT is a classic computer vision algorithm that can
detect distinctive key points and features in images. It's useful for flower recognition, especially
when the dataset is limited.
 4. SURF (Speeded-Up Robust Features): SURF is another feature extraction algorithm that can be
used to recognize flowers in images.
 5. Bag of Visual Words (BoVW): BoVW is a traditional computer vision technique that can be
applied for flower recognition by representing images using histograms of visual words.
 6. Support Vector Machines (SVM): SVMs can be used with handcrafted features or extracted
features from CNNs for flower classification.
ALGORITHMS
 7. Random Forest: Random Forest is an ensemble learning algorithm that can be used for flower
recognition, especially when combining it with appropriate feature extraction methods.
 8. K-Nearest Neighbors (KNN): KNN is a simple but effective algorithm that can be used for
flower recognition based on similarity to neighboring data points.
 9. Principal Component Analysis (PCA): PCA can be used for dimensionality reduction when
working with flower image features, especially in cases with a high number of dimensions.The
choice of algorithm depends on factors like the size and quality of your dataset, the available
computational resources, and the specific requirements of your flower recognition task. In many
cases, deep learning approaches using CNNs and transfer learning tend to provide the best results,
especially when working with large, diverse datasets.
CONCLUSION :
 The development of the flower recognition system represents a significant achievement, successfully
meeting the project's core objectives.
 This system, designed to identify diverse flower species based on visual characteristics, showcases
versatility across various applications.
 Looking ahead, future work involves the continuous refinement of the deep learning model to
enhance recognition accuracy.
 The inclusion of an even broader spectrum of flower species and variations will make the system an
invaluable resource for research and conservation in diverse ecosystems.
 Moreover, expanding partnerships with conservation organizations and researchers will amplify the
system's role in biodiversity monitoring and preservation, further cementing its place at the
intersection of technology and the natural world.

Flower Recog System

Uploaded by

Copyright:

Available Formats

Flower Recog System

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Flower Recog System

Uploaded by

Copyright:

Available Formats

A MINI PROJECT PRESENTATION

DATTA MEGHE COLLEGE OF ENGINEERING

REQUIREMENT AND ANALYSIS

• Excel sheet for Flower data

You might also like