Get in touch
or send us a question?

Traffic sign recognition – 03 – algorithms in object detection

1. What is object detection?

Object detection is a common word to describe a group of mission of computer vision. There are 3 basic missions:

  • Image classification: predict label of object in a picture
  • Object location: find the location of object in a picture and return them by bounding box, that contains

coordinates, width, height

  • Detect object: label to each bounding box

2. Algorithms

  1. R-CNN (2014)

First introduced at 2014 by Ross Girshick and colleagues. There are 3 main components:

  • Region proposal: To create and export the areas that contains object border by bounding box
  • Feature Extractor: Export main features from region proposal through deep convolutional neural network.
  • Classifier: From features, it classify image in region proposal into correct label