Image Retrieval and Classification Techniques

A hybrid geometric-feature algorithm for 2D shape similarity

This paper develops a hybrid algorithm for 2D shape similarity that outperforms geometric and feature-based methods by combining global and local features without requiring training data.

Local Mean Gradient Pattern (LMGP): A novel approach made for the classification of brain CT scan images

The paper introduces LMGP, a feature extraction method for classifying brain CT scans into normal, ischemic, or hemorrhage categories, achieving high accuracy with RBF-SVM.

Plant leaf tooth feature extraction

This study proposes an image-based method to extract and measure leaf tooth features for plant classification and evolutionary analysis, achieving high accuracy.

Image analysis for taxonomic identification of Javanese butterflies

The paper evaluates image analysis methods for classifying Javanese butterflies and slipper orchids, finding that BGR-SURF achieved the best accuracy for butterflies while the original BGR method was best for orchids.