A

Project Report
On
"Bottle Segregation using Live Image Classification"

Prepared by
Kishan Prajapati (20CS065)
Pruthvi Raj (20CS068)

Under the guidance of

Asst. Prof. Vaibhavi Patel

Submitted to
Charotar University of Science & Technology
Degree of Bachelor of Technology
in Computer Science & Engineering
CS452 : Software Group Project-V
of 7th Semester of B.Tech
Submitted at

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING

Faculty of Technology & Engineering, CHARUSAT
Chandubhai S. Patel Institute of Technology
At: Changa, Dist: Anand – 388421
October 2023
 CERTIFICATE

This is to certify that the report entitled “Bottle Segregation using Live Image
Classification” is a bonafied work carried out by Pruthvi Raj (20CS068) under the
guidance and supervision of Asst. Prof. Vaibhavi Patel for the subject Software
Group Project - V (CS452) of 7th Semester of Bachelor of Technology in
Computer Science & Engineering at Faculty of Technology & Engineering
(C.S.P.I.T.) – CHARUSAT, Gujarat.

To the best of my knowledge and belief, this work embodies the work of candidate
himself, has duly been completed, and fulfills the requirement of the ordinance
relating to the B.Tech. Degree of the University and is up to the standard in respect
of content, presentation and language for being referred to the examiner.

Under the supervision of,

Vaibhavi Patel
Asst. Prof.
Dept. of Computer Science & Engineering
C.S.P.I.T., CHARUSAT-Changa.

Dr. Amit Thakkar

Head,
Dept. of Computer Science & Engineering
C.S.P.I.T., CHARUSAT-Changa.

Chandubhai S Patel Institute of Technology (C.S.P.I.T.)

Faculty of Technology & Engineering, CHARUSAT
At: Changa, Ta. Petlad, Dist. Anand, PIN: 388 421. Gujarat
 CERTIFICATE

This is to certify that the report entitled “Bottle Segregation using Live Image
Classification” is a bonafied work carried out by Kishan Prajapati (20CS065)
under the guidance and supervision of Asst. Prof. Vaibhavi Patel for the subject
Software Group Project - V (CS452) of 7th Semester of Bachelor of Technology in
Computer Science & Engineering at Faculty of Technology & Engineering
(C.S.P.I.T.) – CHARUSAT, Gujarat.

To the best of my knowledge and belief, this work embodies the work of candidate
himself, has duly been completed, and fulfills the requirement of the ordinance
relating to the B.Tech. Degree of the University and is up to the standard in respect
of content, presentation and language for being referred to the examiner.

Under the supervision of,

Vaibhavi Patel
Asst. Prof.
Dept. of Computer Science & Engineering
C.S.P.I.T., CHARUSAT-Changa.

Dr. Amit Thakkar

Head,
Dept. of Computer Science & Engineering
C.S.P.I.T., CHARUSAT-Changa.

Chandubhai S Patel Institute of Technology (C.S.P.I.T.)

Faculty of Technology & Engineering, CHARUSAT
At: Changa, Ta. Petlad, Dist. Anand, PIN: 388 421. Gujarat
 DECLARATION BY THE CANDIDATES

We hereby declare that the project report entitled “Bottle Segregation using Live Image
Classification” submitted by us to Chandubhai S. Patel Institute of Technology, Changa in
partial fulfilment of the requirements for the award of the degree of B.Tech Computer
Science & Engineering, from Department of Computer Science & Engineering, CSPIT,
FTE, is a record of bonafide CS452 Software group Project-V carried out by us under the
guidance of Asst. Prof. Vaibhavi Patel. We further declare that the work carried out and
documented in this project report has not been submitted anywhere else either in part or in
full and it is the original work, for the award of any other degree or diploma in this institute or
any other institute or university.

Signature of the candidate

(Kishan Prajapati -20CS065)

Signature of the candidate

(Pruthvi Raj -20CS068)

This is to certify that the above statement made by the candidate is correct to the best
of my knowledge.

Vaibhavi Patel
Asst. Prof.
Department of Computer Science & Engineering,
Chandubhai S Patel Institute of Technology (CSPIT)
Faculty of Technology (FTE)
Charotar University of Science and Technology (CHARUSAT) - Changa.
 Acknowledgement

We are privileged to have this opportunity to express our gratitude and acknowledge
everyone’s never ending support and valuable contributions for our project.
Prima facie, we would like to express my sincere thanks and gratitude to my Lab in charge
Asst. Prof. Vaibhavi Patel , for the continuous support of our project study and related research,
for her patience, motivation, and immense knowledge
Our sincere thanks also goes to HOD Dr. Amit Thakkar who provided us an opportunity to
work on this project and be able to present the same.
Last but not the least, I would like to thank my friends and family for supporting us throughout
this project and for always being a constant source of inspiration. We also place our sense of
gratitude to everyone who supported us while development of this project and lent their hand
in the project.

i
 Abstract

The "Bottle Segregation Using Live Image Classification" project leverages machine learning
and computer vision techniques to create an intelligent system for real-time bottle segregation.
With the rise in environmental concerns and recycling requirements, efficient sorting of bottles
based on their material composition is crucial. This project employs Python, alongside machine
learning libraries, to develop a live image classification model that can instantly identify and
categorize bottles as plastic, glass, or other materials. The system can be integrated with live
camera feed and can provides accurate, real-time segregation, reducing manual labour and
improving the recycling process's efficiency. This innovative solution demonstrates the
potential of machine learning in waste management and contributes to sustainable
environmental practices.

ii
 Table of Contents

Acknowledgement…………………………………………………………………... i
Abstract………………………………………………………………………………ii
Chapter 1 Introduction……………………………………………………………...1
1.1 Overview of Project.............................................................................................. 1
1.2 Objective .............................................................................................................. 1
1.3 Scope .................................................................................................................... 1
1.4 Roles and Responsibilities.................................................................................... 2
1.5 Project plan (Week wise) ..................................................................................... 2
Chapter 2 System Analysis…………………………………………………………. 3
2.1 Study of existing system & its limitations............................................................ 3
2.2 Requirement of new system ................................................................................. 3
2.2.1 Functional Requirements……………………………………………………...3
2.2.2 Non-functional Requirements ………………………………………………...3
2.3 Hardware Requirements ....................................................................................... 4
2.4 Software Requirements ........................................................................................ 4
Chapter 3 System Design……………...……………………………………………...5
3.1 Modules ................................................................................................................ 5
3.2 System flow .......................................................................................................... 5
Chapter 4 Implementation Screenshots……………………………………………7
Chapter 5 Test Cases………………………………………………………………..10
Chapter 6 Limitations and Future Work……………………………………...…..11
Chapter 7 Conclusion……………………………………………………………….12
References……………………………………………………………………………13

iii
 List of Figures

Fig 3.1 System flow ................................................................................................ 6

Fig 4.1.1 Bottle Categories..................................................................................... 7
Fig 4.1.2 Software Interface .................................................................................. 8
Fig 4.1.3 Bottle prediction ..................................................................................... 9
Fig 4.1.4 Captured dataset .................................................................................... 9

iv
 List of Tables
Table 1.1 Roles and Responsibilities .................................................................... 2
Table 1.2 Project plan ............................................................................................ 2
Table 2.1 Test cases .............................................................................................. 10

v
20CS065 CHAPTER1: INTRODUCTION

1 INTRODUCTION
1.1 OVERVIEW OF PROJECT

The "Bottle Segregation Using Live Image Classification" project represents a pioneering
application of machine learning and computer vision to address the pressing issue of
efficient bottle sorting and recycling. In a world increasingly concerned with sustainability
and environmental conservation, the need for precise, real-time identification and
segregation of bottles by material type is paramount.

1.2 PROJECT OBJECTIVE

The project has the potential to improve the overall efficiency of production or recycling
process by real time image classification of different bottles by integrating it with live
video feed to reduce labour and precise bottle sorting.

1.3 PROJECT SCOPE

The scope of the above project is to develop a software system that will be able to identify
different type of bottles in live mode. The system will require the use of advanced python
libraries to interpret the image for identification and machine learning algorithm for
training the model.

The project scope will include the following:

• Development of the software system for image classification model
development
• Integrating system with live camera feed.
• Identification of different bottles based on model trained with live images
captured and dataset
• Material Categories
• User interface design and user experience optimization.

CSPIT 1 DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

20CS065 CHAPTER1: INTRODUCTION

1.4 ROLES AND RESPONSIBILITIES

Kishan (20CS065) Data preprocessing and algorithm implementation

Bugs fixing

Pruthvi (20CS068) Frontend and Camera module

Bugs fixing

Table: 1.1

1.5 PROJECT PLAN (WEEK WISE)

Name of Topic/Module
Week 1 Group formation and project definition finalize

Week 2 Deciding project features

Week 3 Market study of similar platforms.

Week 4 Dataset explore

Week 5 OpenCV implementation

Week 6 Frontend development

Week 7 Data preprocessing
Week 8 Machine learning algorithm implementation
Week 9 Linear SVM
Week 10 Handling errors
Week 11 Work on model prediction
Week 12 Implemented Auto Prediction feature
Week 13 Trying different combinations of predefined dataset and captured dataset
Week 14 Project concluded
Table: 1.2

CSPIT 2 DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

20CS065 CHAPTER 2: SYSTEM ANALYSIS

2 SYSTEM ANALYSIS

2.1 STUDY OF EXISTING SYSTEM AND ITS LIMITATIONS

The current bottle sorting systems heavily rely on manual labor, resulting in inefficiency,
imprecise sorting, high operational costs, and limited scalability. These limitations
underscore the need for an automated, machine learning-based solution to improve the
accuracy and efficiency of bottle segregation while reducing resource intensity and
environmental impact.

2.2 REQUIREMENT OF A NEW SYSTEM

The various limitations of other applications needed to be solved hence this software aims
to implement real time image classification of different types of bottles which will be
helpful to increase the efficiency in the industry.

2.2.1 Functional requirements

- Collecting real time data of different bottles
- Training model using pre-defined dataset as well as real time data
- Preprocessing dataset
- Different bottles identification
- Incorporating bottle identification with live camera feed
- Auto prediction
- Correct image prediction even with limited dataset

2.2.2 Non-functional requirements

- Clean UI.
- Easy navigation.
- User should be easily able to interact with system.
- Good color theme

CSPIT 3 DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

20CS065 CHAPTER 2: SYSTEM ANALYSIS

2.3 HARDWARE REQUIREMENTS

Laptop/PC:
Processor: Intel dual core or above
Processor Speed: 1.0GHZ or above
RAM: 4 GB RAM or above
Hard Disk: 20 GB hard disk or above

2.4 SOFTWARE REQUIREMENTS

Any basic Computer/Laptop with Windows 7/8/10/11 or any recent version of Linux/Mac
OS.

CSPIT 4 DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

20CS065 CHAPTER3: SYSTEM DESIGN

3 SYSTEM DESIGN
3.1 MODULES

There software mainly consists of two modules which are as follows:

Module 1: Dataset gathering and preprocessing

The first module is responsible for the dataset generation. The dataset used in this project
will be image files which can be captured live during the working of project or else the
predefined dataset can also be used which contains images of different types of bottles

Module 2: Image prediction

The second module is responsible for the prediction of bottle based on the categories
defined during testing using the applied machine learning algorithm.

3.2 SYSTEM FLOW DIAGRAM

Components:
Bottle Categories – Define categories of different bottles which will be used for the
project.
Algorithm – Linear SVM for correct model prediction even when the dataset is small.
Model Training – Training the model based on images captured using live camera or
predefined dataset.
Bottle Prediction – The correct prediction of bottle category is done when the system
is shown an image or bottle.

CSPIT 5 DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

20CS065 CHAPTER 3: SYSTEM DESIGN

Fig: 3.1 System flow

CSPIT 6 DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

20CS065 CHAPTER 4: IMPLEMENTATION SCREENSHOT

4 IMPLEMENTATION SCREENSHOTS

Fig: 4.1.1 Bottle categories

CSPIT 7 DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

20CS065 CHAPTER 4: IMPLEMENTATION SCREENSHOT

Fig: 4.1.2 Software Interface

CSPIT 8 DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

20CS065 CHAPTER 4: IMPLEMENTATION SCREENSHOT

Fig: 4.1.3 Bottle prediction

Fig: 4.1.4 Captured dataset

CSPIT 9 DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

20CS065 CHAPTER 5: TEST CASES

5 TEST CASES
Test case Test Pre- Testing Testing Expect Post Status(Pass/F
scenarios case conditio Steps Data ed Condition ail)
Name n Result
Camera Video Camera Run Categori Error Error and fail
availability feed occupied Softwar es system exit
e defined
Camera Video Camera Run Categori Main System UI pass
availability feed not Softwar es system
occupied e defined page

Training Model No Training No Model Error fail

data training training Data not training not
data captured data for trained
model
Training Model Training Different Bottles Model Model pass
data training data images images trained trained
captured of messag message
bottles e
captured
Prediction Bottle Model Show Image of Bottle Bottle pass
identific trained bottle bottle categor category
ation properly for y identified
identific identifi
ation to ed
camera

Table 2.1

CSPIT 10 DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

20CS065 CHAPTER 6: LIMITATIONS

6 LIMITATIONS AND FUTURE WORK

The system has been developed as per the current requirements and include every
necessary feature which was needed.

Limitations
- Can’t guarantee the success rate at basic level
- Limited to three bottle categories for now
- Auto prediction results may fluctuate

Future work
- Integrate it with live camera feed in industry
- Use it for bottle disposal based on different type of bottles

CSPIT 11 DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

20CS065 CHAPTER 7: CONCLUSION

7 CONCLUSION

The "Bottle Segregation Using Live Image Classification" project concludes with the
successful development of an automated system that significantly improves the
efficiency and accuracy of bottle sorting in recycling facilities. By integrating machine
learning and real-time image classification, it not only streamlines the process but also
promotes sustainable waste management. This innovation holds great potential for
reducing operational costs, enhancing recycling quality, and supporting environmental
conservation efforts.

CSPIT 12 DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

20CS065 REFERENCES

REFERENCES

Web Reference:

[1] https://www.python.org/doc/
[2] https://docs.python.org/3/library/tkinter.html
[3] https://scikit-learn.org/stable/modules/svm.html https://devdocs.io/node/
[4] https://opencv.org/
[5] https://stackoverflow.com/

CSPIT 13 DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING