• 0945 09 09 08
  • support@hspace.vn

Traffic Sign Recognition

Course time: Duration: 60
Duration 60 Credits
Categories Projects, Computer Vision

The goal of this course is design an application that can determine the type of the traffic sign that is displayed in an image, and is robust to different real-life conditions such as poor lighting, obstructions or the sign being far away.

Part 1: Concepts, Basic Operations

  • What is Computer Vision?
  • Installation of Python 3 in Windows
  • Installation of OpenCV & Other Packages for Image Processing
  • Setting up the Working Directory for OpenCV
  • Read an Image using OpenCV
  • Save Image using waitKey Function in OpenCV
  • Computer Vision - Basic Concepts of an Image

Part 2 :  Basic Image Operations

  • Basic Image Operations
  • Color Spaces in Image Processing 
  • Properties & Pixels of an Image - Basic Image Operations
  • Draw Geometric Shapes using OpenCV
  • Change Color Space of an Image in OpenCV
  • Basic Image Operations - Splitting Image Channels
  • Plot an Image in Python using matplotlib
  • Plot Multiple Images in Python using matplotlib
  • Region of Interest in an Image - Basic Image Operations using OpenCV

Part 3: Feature Engineering

  •  How to calculate Histogram of Oriented Gradients?
  • Compute the HOG descriptor by using scikit-image library
  • Visualizing HOG 

Part 4: Sliding windowns, Non-Maximum Suppression

  • What is a sliding window?
  • Non-Maximum Suppression
  • Sliding Windows for Object Detection
  • Utilizing non-maximum suppression to ignore redundant, overlapping bounding boxes.

Part 5: Selective Search

  • Image Segmentation
  • How Image Segmentation Works
  • How Selective Search Works
  • How to use Selective Search based segmentation implemented in OpenCV

Part 6: Support Vector Machine

  • What is a SVM?
  • How is the optimal hyperplane computed?
  • Set up the training data
  • Set up SVM's parameters
  • Train the SVM
  • Regions classified by the SVM

Part 7: Traffic sign recognition

  • Traffic sign recognition algorithm
  • Train SVM Traffic sign model
  • Region Proposal by Selective Search
  • Classification of traffic signs based on the svm model and proposed regions

 

hSpace Team

hSpace Team

hSpace Team

Traffic Sign Recognition