This project focuses on implementing real-time object detection using Raspberry Pi and OpenCV. Real-time object detection is a critical aspect of computer vision applications, allowing systems to identify and locate objects within a live video stream instantly.
Report
Share
Report
Share
1 of 10
More Related Content
Real Time Object Detection Using Open CV
1. PROJECT- 1
EC- 711
REAL TIME OBJECT DETECTION
USING RASPBERRY PI & OPEN CV
JAWAHARLAL NEHRU GOVERNMENT ENGINEERING COLLEGE, SUNDERNAGAR
DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGINEERING
Submitted By:
Khem Singh (20010104029)
Reva (20010104044)
Project guide-
Asst. Prof. Pooja Sharma
2. INTRODUCTION
● This project focuses on real-time object detection, a crucial component of
computer vision applications.
● It involves identifying and locating objects within a video stream in real-time.
● Real-time object detection has diverse applications ranging from security and
surveillance to automation and robotics.
● Raspberry Pi serves as the hardware backbone, providing a cost-effective and
versatile platform for deploying our computer vision solution.
3. OBJECTIVE
● Achieve accurate and swift identification of objects within a live video stream.
● Prioritize the real-time aspect, ensuring minimal latency between object
detection and system response.
● Enhance user experience through intelligent automation.
● Monitor and secure areas in real-time, alerting to potential threats or unusual
activities.
4. SYSTEM OVERVIEW
Raspberry Pi:
● The central computing unit that serves as the brains of the system.
● Executes the real-time object detection algorithm, manages system resources, and
facilitates communication between components.
Camera:
● Captures live video feed, providing input for real-time object detection.
● Essential for gathering visual data from the environment .
5. SYSTEM OVERVIEW
OpenCV Library:
● Provides a robust set of tools and functions for image and video processing.
● Implemented in Python code running on Raspberry Pi, enabling seamless
interaction with the camera feed.
6. SYSTEM OVERVIEW
Pre-trained Model:
● Utilizes a pre-trained deep learning model for object detection .
● SSD (Single Shot Multibox Detector) with MobileNetV3 architecture .
7. OVERVIEW OF DATASET
● Data set used- “COCO (Common Objects in Context) ".
● The COCO dataset is a large-scale object detection, segmentation, and
captioning dataset
● It includes 91 object categories, providing a diverse and challenging set of
images.
● Some common object labels in the COCO dataset include person, bicycle,
car, airplane, bus, train, truck, bird, cat, dog, horse, etc.