Skip to content

Latest commit

 

History

History
23 lines (19 loc) · 1.08 KB

README.md

File metadata and controls

23 lines (19 loc) · 1.08 KB

Image classification steps

This project contains the necessary code and images to understand the image classification process. Most images are in the image folder (Day and night to be added), see below for the order of the files, each individual file contains code and instructions.

Project files in order are as follows:

  • Images as Numerical Data (Images as a grid of pixels)
  • Visualising RGB channel (RGB colourspace)
  • Pre-processing Examples (Cropping and resizing)
  • Green Screen car (Colour Masking, Green screen)
  • Colour Conversions (HSV colour conversion) - (Work in progress)
  • Visualising the data (Day and night image classifier)
  • Standardising the data (Day and night image classifier)
  • Average Brightness feature extraction (Day and night image classifier)
  • Finding edges and custom kernals (Creating a filter, edge detection)
  • Histogram and Feature Vectors (Histograms of colour)
  • Classification - (Work in progress)
  • Accuracy and miscalculation - (Work in progress)

To do:

  • Add day and night images
  • Finish Colour conversion
  • Redo classification to improve accuracy