OBJECT DETECTION
SHAGUN(23010204038)
SHALINI THAKUR(23010204039)
SHABANAM KUMARI(2402060013)
INTRODUCTION:
The object detection system is a surveillance technology designed to detect, track, etc. security
threats within a designated area.
The system consists of Arduino uno, servo motor, ultrasonic sensor, are devices that generate or
sense ultrasound energy. They can be divided into three broad categories: transmitters, receivers
and transceivers.
The Arduino Uno is a microcontroller board that works by reading inputs and turning them into
outputs. It can be programmed to control a variety of electronic components, such as motors,
lights, and buttons.
Arduino Uno can read inputs from sensors, buttons, or other sources. Turn inputs into outputs: The
Arduino Uno can turn inputs into outputs, such as activating a motor, turning on an LED, or
publishing something online.
The servo motor works by receiving a command signal, moving the motor, and then comparing the
actual position to the command signal. If there's a difference, the motor is adjusted to correct the
position. This process is called closed-loop control.
Project objective:
The goal of an object detection project is to develop a computer vision system that can accurately
identify and locate objects within images or videos.
Object Detection Tasks
1. Classification: Detect the type of object (e.g., person, car, dog).
2. Localization: Determine the location of the object within the image or video (e.g., bounding
box).
3. Detection: Combine classification and localization to detect objects.
This overview provides a general outline of an object detection project. Depending on the specific
requirements and goals, the project may involve additional steps, techniques, and considerations.
Object detection is a fundamental task in computer vision, enabling machines to identify and locate
specific objects within images or videos. With smart devices, autonomous vehicles, and security
systems, the demand for efficient and accurate object detection algorithms has never been greater.
Project Overview:
This project, titled "Smart Vision or object detection," aims to develop a robust
and efficient object detection system using deep learning techniques. The system
will be designed to detect and classify objects within images or videos, with
potential applications in various fields, including:
1. Surveillance: Anomaly detection, and monitoring.
2. Autonomous Vehicles: Pedestrian detection, lane detection, and obstacle
avoidance.
Methodology
The project will employ a combination of the following methodologies:
1. Literature review: Studying existing object detection algorithms and
techniques.
2. Dataset collection and annotation: Gathering and annotating a dataset for
training and testing.
3. Model development and training: Developing and training a deep learning.
Expected Outcomes
The expected outcomes of this project include:
1. A robust object detection algorithm: Capable of detecting and classifying objects
with high accuracy and efficiency.
2. A demonstration of potential applications: Showcasing the system's capabilities in
various domains.
3. A comprehensive report and documentation: Detailing the project's methodology,
results, and conclusions.
Component used:
Ultrasonic sensor (V2.0B)
Breadboard
Arduino uno
Servo motor
Jumper wire
Arduino cable
[Link] sensor V2.0B:
The Ultrasonic Sensor V2.0B is typically used for measuring distance, detecting objects, and
avoiding obstacles. It works by emitting ultrasonic waves and measuring the time it takes for the
echo to return after bouncing off an object.
Common use:
Distance Measurement: Used in robotics, automation, and IoT applications to
measure the distance between the sensor and an object.
Obstacle Avoidance: Helps autonomous robots and drones detect and avoid obstacles.
Liquid Level Monitoring: Measures liquid levels in tanks without direct contact.
Security Systems: Used in motion detection for alarms and surveillance.
Parking Assistance: Helps vehicles detect nearby objects to avoid collisions.
[Link] uno:
The Arduino Uno is a popular microcontroller board based on the ATmega328P. It's a versatile
platform for building interactive projects, from simple circuits to complex robots.
Key Features:
1. Microcontroller: ATmega328P with 14 digital input/output pins, 6 analog inputs,
and 16 MHz clock speed.
2. Memory: 32 KB flash memory, 2 KB SRAM, and 1 KB EEPROM.
3. Communication: USB, serial, and I2C interfaces.
4. Power: USB-powered or external 7-12V DC power supply.
Uses of Arduino Uno:
o Embedded Systems Development – Used to create and test embedded applications.
o Robotics – Controls motors, sensors, and actuators in robotics projects.
o Home Automation – Automates lights, fans, doors, and other household devices.
o IoT (Internet of Things) Applications – Connects sensors to the internet for data
collection and remote control.
o Educational Purposes – Teaches students about programming, electronics, and
hardware interfacing.
o DIY Hobby Projects – Used in various creative projects like LED displays, sound
controllers, and gaming devices.
[Link] motor:
A servo motor is a type of motor that provides precise control of angular position, velocity, and
acceleration. It consists of a motor, a position sensor, and a control circuit. Servo motors are widely
used in various applications due to their accuracy and efficiency.
Steps:
Internal working of servo motor
1. Receive a command signal
The servo motor receives a signal that specifies the desired position, velocity, or torque.
2. Move the motor
The motor's internal circuitry powers the motor to move towards the desired position.
3. Monitor the motor's position
A position sensor monitors the actual position of the motor's output shaft.
4. Compare the actual position to the desired position
The control circuit compares the actual position to the desired position.
[Link] wires:
Jumper wires are essential in object detection projects for establishing electrical
connections between various components such as sensors, motors, and microcontrollers
like Arduino. They are commonly used in prototyping circuits on breadboards
Jumper wires create secure connections for:
Arduino Uno
Ultrasonic Sensor (V2,0B)
Breadboard
VCC (Ultrasonic Sensor) → 5V (Arduino) (Red wire)
GND (Ultrasonic Sensor) → GND (Arduino) (Black wire)
TRIG (Ultrasonic Sensor) → Digital Pin 9 (Arduino) (Green wire)
ECHO (Ultrasonic Sensor) → Digital Pin 10 (Arduino) (Yellow wire).
[Link]:
A breadboard is a crucial tool in object detection projects as it allows you to build and
test circuits without soldering. It’s especially useful when working with Arduino,
ATmega328P, and various sensors for object detection.
[Link] cable:
An Arduino cable is essential for connecting your Arduino board to a computer for:
1. Power Supply — Provides 5V power to the board during programming or testing.
2. Code Upload — Transfers your code (e.g., object detection logic) from the Arduino
IDE to the microcontroller.
3. Serial Communication — Enables real-time data monitoring via the Serial Monitor.
Steps to Connect Arduino with Cable
Plug the USB Type-A end into your computer.
Connect the Type-B/Micro USB/Type-C end to your Arduino.
Open the Arduino IDE and select the correct Port and Board Type under Tools.
Upload your object detection code and monitor sensor data via the Serial Monitor.
Simulation:
Project Progress:
Current Status
• The object detection project is currently in the model development and training
phase. We have made significant progress in the past few weeks, and here are some
key updates: Challenges and Solutions.
Object occlusion:
• We observed that some objects were partially occluded by other objects. To handle
this, we used data augmentation techniques, such as rotation and flipping, to
increase the diversity of the dataset.
Next Steps
1. Model evaluation and optimization: We will evaluate the model's performance
on a separate test set and optimize it’s for better accuracy and efficiency.
2. Deployment: enabling real-time object detection and classification.
Conclusion:
The object detection project is progressing well, with significant accomplishments in
dataset collection, model selection, and training. We are addressing challenges related
to class imbalance and object occlusion, and we are on track to complete the project
within the anticipated timeline.