Pyramided and optimized blurred shape model for plant leaf classification

Plant leaf classification is a crucial task in the field of computer vision and pattern recognition, with various applications such as plant species identification disease diagnosis. In this paper, authors introduce Pyramid Blurred Shape Model (PBSM) new descriptor for classification. The PBSM extracts both shape texture features from images, which are combined to define probability density function shape. Our experimental results show that proposed achieves high accuracy, F1-score, precision-recall results, demonstrating its effectiveness However, extracting all available images can lead redundant inessential features, degrade performance computational efficiency. To address issue, implement grey wolf optimization (GWO)-based feature selection identify most informative final set then classified using list selected classifiers, further enhancing authors’ approach. evaluate their method on three publicly datasets, namely Middle European Woods (MEW), Swedish, Flavia achieve accuracies 96.34%, 96.89%, 92.41% Flavia, MEW respectively. approach outperforms state-of-the-art descriptors terms accuracy robustness, potential real-world applications. Overall, provides reliable efficient solution It contribute development automated systems diagnosis, thereby facilitating conservation protection species.