Skip to content

A brand logo recognition system using deep convolutional neural networks.

License

Notifications You must be signed in to change notification settings

ifcoxs/DeepLogo

 
 

Repository files navigation

DeepLogo

A brand logo recognition system using deep convolutional neural networks.

Usage

  1. python crop_and_aug.py: Crop brand logo images from the flickr27_logos_dataset and apply data augmentation method. Finally the dataset consists of 217488 images.
  2. python gen_train_valid_test.py: Generate(Split) train/valid/test set from the dataset.
  3. python train_deep_logo_cnn.py: Train the convolutional neural networks and save the trained model to disk.
  4. python test_deep_logo_cnn.py: Test the trained model.

Network

The network is based on this blog post. Same network is applied to this brand logo recognition task because a brand logo is similar to a number plate which consists of a number of digits and letters.

License

MIT

About

A brand logo recognition system using deep convolutional neural networks.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%