Search icon CANCEL
Arrow left icon
Explore Products
Best Sellers
New Releases
Books
Videos
Audiobooks
Learning Hub
Conferences
Free Learning
Arrow right icon
Arrow up icon
GO TO TOP
Hands-On Vision and Behavior for Self-Driving Cars

You're reading from   Hands-On Vision and Behavior for Self-Driving Cars Explore visual perception, lane detection, and object classification with Python 3 and OpenCV 4

Arrow left icon
Product type Paperback
Published in Oct 2020
Publisher Packt
ISBN-13 9781800203587
Length 374 pages
Edition 1st Edition
Languages
Tools
Arrow right icon
Authors (2):
Arrow left icon
Krishtof Korda Krishtof Korda
Author Profile Icon Krishtof Korda
Krishtof Korda
Luca Venturi Luca Venturi
Author Profile Icon Luca Venturi
Luca Venturi
Arrow right icon
View More author details
Toc

Table of Contents (17) Chapters Close

Preface 1. Section 1: OpenCV and Sensors and Signals
2. Chapter 1: OpenCV Basics and Camera Calibration FREE CHAPTER 3. Chapter 2: Understanding and Working with Signals 4. Chapter 3: Lane Detection 5. Section 2: Improving How the Self-Driving Car Works with Deep Learning and Neural Networks
6. Chapter 4: Deep Learning with Neural Networks 7. Chapter 5: Deep Learning Workflow 8. Chapter 6: Improving Your Neural Network 9. Chapter 7: Detecting Pedestrians and Traffic Lights 10. Chapter 8: Behavioral Cloning 11. Chapter 9: Semantic Segmentation 12. Section 3: Mapping and Controls
13. Chapter 10: Steering, Throttle, and Brake Control 14. Chapter 11: Mapping Our Environments 15. Assessments 16. Other Books You May Enjoy

Chapter 7

  1. SSD is a neural network able to find multiple objects in an image, and its output includes the position of the object detected. It can work in real time.
  2. Inception is an influential and precise neural network created by Google.
  3. A layer that has been frozen cannot be trained.
  4. No, it cannot. It can only detect traffic lights, but not their color.
  5. Transfer learning is the process of taking a neural network trained on a task and adapting it to solve a new, related task.
  6. Adding dropout, increasing the size of the dataset, increasing the variety of data augmentation, and adding batch normalization.
  7. Given the variety of images in ImageNet, it's difficult to choose the kernel size of the convolutional layers, so they used kernels of multiple sizes in parallel.
lock icon The rest of the chapter is locked
Register for a free Packt account to unlock a world of extra content!
A free Packt account unlocks extra newsletters, articles, discounted offers, and much more. Start advancing your knowledge today.
Unlock this book and the full library FREE for 7 days
Get unlimited access to 7000+ expert-authored eBooks and videos courses covering every tech area you can think of
Renews at $19.99/month. Cancel anytime