Improve the efficiency of handcrafted features in image retrieval by adding selected feature generating layers of deep convolutional neural networks.
Ghazal ShamsipourShervan Fekri ErshadMahdi SharifiAlireza AlaeiPublished in: Signal Image Video Process. (2024)
Keyphrases
- convolutional neural networks
- image retrieval
- feature set
- image features
- feature vectors
- selected features
- multiple features
- feature interactions
- binary features
- additional features
- feature values
- individual features
- input features
- invariant features
- selecting features
- predictive features
- improve the recognition accuracy
- image similarity
- redundant features
- significant features
- feature weights
- global features
- feature representation
- irrelevant features
- feature analysis
- discriminative features
- low level
- sequential forward selection
- low level visual features
- spatial features
- cbir systems
- retrieval method
- feature subset
- keypoints
- image database