Scribd
Upload
42 views

DL Documentation

The document discusses object detection using the YOLOv5 deep learning algorithm. It provides an overview of YOLOv5 and its methodology for object detection, which involves image preprocessing, using a backbone network to extract features, detecting objects through convolutional layers, non-maximum suppression to remove duplicate detections, confidence scoring, and outputting bounding boxes and labels. YOLOv5 provides state-of-the-art performance on the COCO dataset and can be applied to applications like autonomous vehicles and surveillance.

Uploaded by

rahul
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
42 views

DL Documentation

The document discusses object detection using the YOLOv5 deep learning algorithm. It provides an overview of YOLOv5 and its methodology for object detection, which involves image preprocessing, using a backbone network to extract features, detecting objects through convolutional layers, non-maximum suppression to remove duplicate detections, confidence scoring, and outputting bounding boxes and labels. YOLOv5 provides state-of-the-art performance on the COCO dataset and can be applied to applications like autonomous vehicles and surveillance.

Uploaded by

rahul
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Object Detection using YOLOV5

by 20761A5417

March 12, 2023

1 Abstract

This lookup focuses on object detection the use of the deep mastering algo-

rithm YOLOv5, an up to date model of YOLO that gives multiplied accuracy

and performance. According to the findings, YOLOv5 provides today’s overall

performance on the COCO dataset, with a precision rating of 0.709, with recall

rating 0.634 and with mAP50(mean average precision) 0.713. The learn about

demonstrates the doable of YOLOv5 for sensible applications, inclusive of in-

dependent vehicles, security, and surveillance. In this research, we inspect the

utilization of YOLOv5 deep mastering strategies for object detection.

2 Introduction

Object detection is a crucial trouble in laptop vision, with functions ranging

from surveillance and protection to self-driving cars. Deep studying algorithms

have verified awesome success in fixing object detection problems, and YOLO

(You Only Look Once) is one of the most famous deep gaining knowledge of

object detection methods. YOLOv5, the modern day model of YOLO, has

accelerated overall performance and accuracy. In this research, we look at the

utilization of YOLOv5 for object detection.

1
3 Methodology

The YOLOv5 object detection device includes the following steps:

A. Image Preprocessing: The enter picture is resized to a constant di-

mension and normalized to make sure that it meets the enter necessities of the

YOLOv5 model.

B. Backbone Network: YOLOv5 makes use of a spine community to ex-

tract elements from the enter image. YOLOv5 makes use of the CSPDarknet53

backbone, which is a deeper and wider model of the Darknet53 community used

in YOLOv3.

C. Object Detection: YOLOv5 makes use of a detection head to generate

object detections from the extracted features. The detection head consists of

a set of convolutional layers that predict the type possibilities and bounding

container coordinates for every object.

D. Non-maximum Suppression: Since more than one bounding contain-

ers can also predict the identical object, non-maximum suppression is utilized

to take away replica detections and hold solely the most assured ones.

E. Confidence Score Calculation: The self assurance rating for every de-

tection is calculated by means of multiplying the classification chance and the

Intersection over Union (IoU) between the anticipated bounding field and the

floor reality bounding box. The detections with excessive self assurance ratings

are kept, whilst the others are discarded.

F.Output: The closing output of the YOLOv5 machine is a set of bounding

boxes, category labels, and self assurance rankings for the objects detected in

the enter image.

2
4 Final Result:

3
5 Conclusion:

In conclusion, object detection is a crucial task in computer vision that has

many real-world applications, ranging from surveillance to autonomous driv-

ing. The YOLOv5 algorithm has proven to be a highly effective and efficient

approach for object detection in various scenarios. The algorithm’s ability to

detect objects accurately and quickly has made it a popular choice among re-

searchers and practitioners.

You might also like