Skip to content

Abstract and handcrafted feature fusion scheme for VHR image classification

Notifications You must be signed in to change notification settings

sbAsma/CNN_MSLIC_Haralick_Feature_Fusion

Repository files navigation

CNN and M-SLIC Superpixels Feature Fusion for VHR Image Classification

This is the code used to generate the results of [1]. Please refer to [1] for more details. Please cite [1] if you are to use the dataset.

[1] B. A. Semcheddine and A. Daamouche, "CNN and M-SLIC Superpixels Feature Fusion for VHR Image Classification," 2022 International Conference of Advanced Technology in Electronic and Electrical Engineering (ICATEEE), M'sila, Algeria, 2022, pp. 1-4, doi: 10.1109/ICATEEE57445.2022.10093756.

Note

The SCf and SHCf methods might give slightly different results than what you will find in [1], because of the randomness in the initialization and during the learning process of the CNN. This code has also a much lower number of epochs in the training process of the CNN, prior to the feature extraction phase.

Dependencies

  • python == 3.7
  • numpy == 1.21.6
  • pandas == 1.3.5
  • sklearn == 1.0.2
  • cv2 == 3.4.2
  • matplotlib == 3.1.3
  • skimage == 0.19.2
  • scipy == 1.7.3
  • tensorflow == 2.3.1
  • keras == 2.4.0
  • PIL == 7.0.0

Releases

No releases published

Packages

No packages published