JSPM’s Nahre Technical Campus

Department of Electronics and Telecommunication Engineering

PROJECT BASED ON,

AU T O M AT I C O B J E C T S O R T I N G U S I N G
CONVEYOR BELT

Presented By,
Rushikesh Kurtadkar
Ketaki Lokare
Riddhi Kadam
Sujata Patil
2

CONTENTS
o Introduction o Circuit Diagram

o Literature Review o Relevance

o Problem Formulation o Methodology

o Problem Solution o Flow Chart

o Objectives o Work Plan Schedule

o Block Diagram o References

 I N T RO D U C T I O N
[1] This project integrates an ESP8266 camera module for live video streaming, enabling real-
time monitoring of the sorting process and allowing operators to ensure proper system
functionality. Alongside, an independent image processing algorithm runs on a laptop,
analyzing the live stream to provide deeper insights into the materials being sorted without
interfering with the conveyor's operation. By separating image analysis from the mechanical
sorting process, the system maintains high efficiency while offering enhanced analytical
capabilities, showcasing a practical solution for automated material separation in industrial
applications.

Additionally, the system's scalability allows it to be easily adapted for various production line
sizes and material types, making it versatile for different industrial applications. Its energy-
efficient design also contributes to lowering operational costs while maintaining high
throughput. By integrating advanced sensing technologies and a robust mechanical setup, this
conveyor system not only boosts efficiency but also promotes sustainable practices by
optimizing resource use and minimizing waste, aligning with the growing emphasis on
environmentally friendly manufacturing processes.
L I T E R AT U R E R E V I E W
Comparative Analysis of the Literature Review Papers
5 L I T E R AT U R E PA P E R 1
Aspects Literature Paper Project Comments
Enhancing industrial Sorting airport luggage Both aim at improving efficiency
efficiency by sorting efficiently using image using automation and image
Objective
objects using image recognition for empty spaces processing.
processing
Computer Vision Image recognition Both systems rely on image
Primary (OpenCV), Raspberry Pi 3, SIFT, and system, Conveyor belt, ESP32. processing, but your project focuses on empty
Technology Used SVM customized sorting algorithms space detection.

SIFT for feature Likely to use custom or While the literature uses SIFT for
Image
extraction, SVM for machine learning general object classification your project would
Processing
classification algorithms for space and require algorithms for detecting empty spaces and
Algorithm
luggage detection specific luggage types.
Raspberry Pi 3 ESP32 ESP32.
Controller emphasizing its reliability in real-
time systems.
Real-time object Real-time luggage Performance metrics such as fps
Object Detection recognition at 10 fps recognition possibly could be tested and compared to
Method comparable fps assess efficiency.
depending on hardware
Motors and Stepper and Servo Conveyor belt system Both use motors, but the
Mechanical Motors (possibly with mechanical complexity may vary based on project
Structure Stepper/Servo motors) requirements.
General industrial Airport luggage sorting Your project targets a specific application area ,
object sorting which requires specialized image
Application Area
recognition.
6 L I T E R AT U R E PA P E R 2
Aspects Literature Paper Project Comments
Efficient sorting of objects in small Reducing human labor in airport Both projects emphasize automation to reduce
Objective and medium-scale industries luggage sorting while minimizing manual labor, but your project targets a more
to reduce manual labor luggage damage specific and sensitive application: airport
MATLAB for image processing RGB ESP32, image recognition for Your system uses real-time image processing and
Primary
color models. Binary transformation luggage detection, object detection, which may need to handle more
Technology
for object detection conveyor belt system complex variables like luggage size, shape, and
Used
empty spaces.
RGB color model and binary Real-time image recognition (possibly Both projects use image processing but focus on
Image
transformation to distinguish objects using OpenCV) to detect luggage and different features: color vs. space/luggage
Processing
by color empty spaces detection.
Method
Atmega16 ESP32 Your project benefits from the processing power
Controller microcontroller and versatility of the Raspberry Pi, whereas the
literature review uses a simpler microcontroller.
Minimizes human intervention in Reduces human intervention in Both systems reduce labor costs, but your project
Automation object sorting to improve productivity airports, minimizing errors and adds the dimension of handling delicate and often
Impact potential damage heavy items like luggage
MATLAB for image OpenCV and possibly MATLAB is more commonly used for academic
Software processing other Python libraries and research purposes, while OpenCV with
Utilized with ESP32 Raspberry Pi is a practical choice for real-world
implementations.
Small and medium-scale industrial Airport luggage sorting Your project is applied in a larger- scale, more
Application sorting critical environment where efficiency and accuracy
Area are paramount.
 P R O B L E M F O R M U L AT I O N

Labor-Intensive Complexity of Scalability Environmental Real Time Data

Processes Pneumatic Systems Challenges Concerns Management

Current sorting Pneumatic systems As production Inefficient sorting Lack of integrated

processes often rely are complex, costly, volumes grow, leads to more waste, data management
on manual labor, and require regular traditional sorting raising environmental systems results in
leading to maintenance. methods struggle to concerns. limited visibility and
inefficiencies, human keep up with the control over the
error, and increased demand for speed sorting process,
operational costs. and accuracy. hindering real time
decision
 P RO B L E M S O L U T I O N
Multidisciplinary
Automated Sorting Integration of Defective Item Cloud Based Data
Integration and
with Conveyor Belt Image Recognition Identify Storage and Access
Scale
Implement an Use advanced image Integrate a detection Store sorting data in Combine elements of
automated conveyor recognition mechanism that the cloud, allowing mechanical
belt system that technology to automatically for real time engineering,
reduces the reliance accurately identify identifies defective monitoring, analysis, computer vision, and
on manual labor, and sort objects items and redirects and accessibility cloud computing to
enhancing efficiency based on parameters them to recycling or through a user create a scalable and
and reducing the like height, width, discarding sections, friendly interface. adaptable solution
potential for human weight, and design maintaining quality This facilitates better that can be
error in the sorting layout, ensuring high control. [4] decision making and customized for
process. precision in sorting. process optimization. various industries and
production scales.
9 OBJECTIVES
1. Automate Material Separation: Use an inductive proximity sensor for efficient metal and non-metal sorting.
2. Utilize Cost-Effective Components: Implement a 12V DC motor to reduce costs and complexity, making the
system viable for small to medium-sized enterprises.
3. Integrate Real-Time Monitoring: Use an ESP8266 camera for live-stream monitoring of the sorting process.
4. Enable Independent Image Processing: Run an image processing algorithm to analyze materials
independently from the conveyor's operation.
5. Enhance Operational Efficiency: Streamline the sorting process to increase throughput and improve
productivity.
6. Improve Data Collection and Analysis: Gather and analyze data to optimize sorting and support informed
decision-making.
7. Promote Sustainability: Improve material recovery efficiency to minimize waste and support environmental
sustainability.
8. Demonstrate Scalability and Adaptability: Create a system that can be easily scaled and adapted for various
industries, including recycling and manufacturing.
10 BLOCK DIAGRAM

ESP 32 CAM LIVE STREAMING

ESP 8266 CONVEYOR BELT

INDUCTIVE
L298 MOTOR DRIVER DC MOTOR
PROXIMITY SENSOR
11 C I RC U I T D I A G R A M
12 R E L E VA N C E
1. Increased Efficiency: Automated sorting systems enhance productivity by speeding up material processing and
reducing manual labor.
2. Cost-Effectiveness: Utilizing a 12V DC motor instead of pneumatic systems lowers operational costs, making
the solution accessible for smaller enterprises.
3. Improved Accuracy: Inductive proximity sensors increase sorting precision, crucial for maintaining high-quality
standards in recycling operations.
4. Real-Time Monitoring: The ESP8266 camera module enables live monitoring, allowing operators to quickly
address issues and maintain smooth operations.
5. Independent Analysis: Separating image processing from the sorting mechanism allows for advanced material
analysis without disrupting the sorting process.
6. Sustainability: The project promotes sustainable practices by improving material recovery efficiency and
reducing waste sent to landfills.
7. Adaptability: The system design can be easily modified for various applications across different industries,
enhancing sorting processes in multiple sectors.
 METHODOLOGY
Automatic object sorting using real-time image
recognition, conveyor systems and Raspberry Pi for
efficient and damage-free handling.
14

1 . R E Q U I R E M E N T S G AT H E R I N G & S Y S T E M
DESIGN
o Objectives : o Activities :
Define the specific requirements for Conduct stakeholder interviews to
the system, including the types of understand the operational
objects to be sorted, the parameters environment and specific needs.
for sorting (e.g., height, width, Identify hardware and software
weight, design), defect detection requirements, including conveyor belt
criteria, and data management specifications, sensors, cameras, and
needs. computing resources.
Design the system architecture,
integrating mechanical, electrical, and
software components.
15

2.HAR DWAR E S E LE C T ION & M E C HANIC AL

SETUP
o Objectives : o Activities :
Select and assemble the mechanical Choose appropriate motors, belts
and electrical components of the and drive systems based on the load
conveyor belt system. and speed requirements. Integrate
sensors to detect object parameters.
Set up high resolution cameras for
image capture, ensuring they are
positioned to cover the entire
conveyor area.
Implement a mechanical mechanism
for sorting and diverting objects
based on the classification results.
16

3.IMAGE RECOGNITION SYSTEM

D E P L OY M E N T
o Objectives : o Activities :
Develop the computer vision system Train machine learning models using
to accurately identify and classify labeled datasets to recognize objects
objects based on the predefined based on their height, width, weight,
parameters. and design layout.
Develop algorithms for real time
object detection and defect
identification.
Test and finetune the image
recognition models to ensure high
accuracy and robustness, including
edge case scenarios.
17

4.DEFECT DETECTION AND SORTING

L O G I C I M P L E M E N TA T I O N
o Objectives : o Activities :
Implement the logic for identifying Develop algorithms to detect defects
defective objects and triggering based on visual characteristics or
appropriate sorting actions. deviations from standard dimensions.
Program the sorting logic to classify
objects into different categories (e.g.,
pass, recycle, discard) and control the
mechanical sorting mechanisms
accordingly.
Simulate and test the sorting process
in different scenarios to ensure
reliability and efficiency.
18

5 . S Y S T E M I N T E G R AT I O N A N D I T ’ S
TESTING
o Objectives : o Activities :
Integrate all system components and Integrate the mechanical, electrical,
perform comprehensive testing to and software components, ensuring
ensure seamless operation. smooth communication and
Perform system-wide integration coordination between systems.
tests to simulate real-world Conduct unit tests on individual
operations, identifying and resolving components (e.g., conveyor belt
any issues. operation, image recognition
accuracy, sorting mechanism) to
validate functionality.
19

6 . S Y S T E M O P T I M I Z AT I O N A N D
C A L I B R AT I O N
o Objectives : o Activities :
Optimize the system for maximum Fine-tune the conveyor belt speed,
efficiency, accuracy, and reliability. camera settings, and sorting
mechanisms for optimal
performance.
Calibrate the image recognition
models and sensors to account for
variations in object types and
environmental conditions.
Implement feedback loops and
adaptive algorithms to improve
sorting accuracy over time.
20

7 . S Y S T E M D E P L OY M E N T A N D
MONITORING
o Objectives : o Activities :
Deploy the system in the target Install the system in the operational
environment and establish environment, ensuring proper
monitoring protocols to ensure alignment and setup.
continued performance. Train personnel on system operation,
Establish a maintenance schedule maintenance, and troubleshooting.
and support plan to address any Set up monitoring tools to track
issues that arise during operation. system performance, including
sorting accuracy, processing speed,
and system uptime.
21

8 . S Y S T E M D O C U M E N TA T I O N A N D
REPORTING
o Objectives : o Activities :
Document the entire development Prepare detailed documentation for
process, including design decisions, system architecture, software
test results, and user manuals. algorithms, and hardware
configurations.
Create user manuals and training
materials for operators and
maintenance personnel.
Compile a final report summarizing
the project outcomes, including
performance metrics and future
improvements.
22 FLOW CHART
 WORK PLAN AND
SCHEDULE
Structured timeline for implementing real-time image
recognition and conveyor-based luggage sorting
system, from design to testing and deployment.
24
WORK PLAN AND SCHEDULE
Phase Objective Key Activities Duration

- Stakeholder interviews
1. Requirements Gathering and System
Define system requirements. 2 Weeks
Design - Identify requirements

- Design architecture
- Choose motors and belts

2. Hardware Selection and Mechanical Assemble mechanical and - Integrate sensors and cameras
3 Weeks
Setup electrical components.
- Set up sorting mechanisms

- Train models
3. Image Recognition System Develop and integrate image - Develop algorithms 2 Weeks
Development recognition.
- Test integration
25
WORK PLAN AND SCHEDULE
Phase Objective Key Activities Duration

- Develop detection algorithms

4. Defect Detection and Sorting Logic Implement defect detection and
- Program sorting logic 2 Weeks
Implementation sorting logic.

- Simulate and test

- Develop database

5. Data Management and Cloud Implement data management and - Implement synchronization
2 Weeks
Integration cloud storage.
- Create interface
- Ensure data security

- Integrate components
6. System Integration and Testing Integrate and test all components. 2 Weeks
- Conduct tests
- Resolve issues
26
WORK PLAN AND SCHEDULE
Phase Objective Key Activities Duration

- Fine-tune components

7. Optimization and Calibration Optimize system performance. - Calibrate models 1.5 Weeks

- Implement feedback loops

- Install system

Deploy system and establish - Train personnel

8. Deployment and Monitoring 2 Weeks
monitoring.
- Set up monitoring and maintenance

- Prepare documentation
Document process and report
9. Documentation and Reporting - Create manuals 2 Weeks
outcomes.
- Compile final report
27
REFERENCES
[1] https://www.researchgate.net/publication/329056573_Object_Sorting_using_Image_Processing
[2] https://www.academia.edu/63343524/Sorting_of_Objects_Using_Image_Processing
[3] https://www.ijnrd.org/papers/IJNRD1803005.pdf
[4] https://youtu.be/zfiSAzpy9NM?si=9SzdHGjL4LLK7RGX
[5] https://www.semanticscholar.org/paper/Sorting-of-Objects-Using-Image-Processing-Yadav-
Uikey/98ec44a400d16d510cb178453347d67db1d958c2#paper-topics
T H A N K YO U
Rushikesh Kurtadkar
Ketaki Lokare
Riddhi Kadam
Sujata Patil