%0 Generic %1 M.2008 %A Nilsback, M. %A Zisserman, Andrew %D 2008 %J 2008 Sixth Indian Conference on Computer Vision, Graphics & Image Processing %K flower_classification machine_learning computer_vision Oxford102_flower_dataset posted_with_chatgpt %P 722-729 %R 10.1109/ICVGIP.2008.47 %T Automated Flower Classification over a Large Number of Classes %U https://www.semanticscholar.org/paper/02b28f3b71138a06e40dbd614abf8568420ae183 %X We investigate to what extent combinations of features can improve classification performance on a large dataset of similar classes. To this end we introduce a 103 class flower dataset. We compute four different features for the flowers, each describing different aspects, namely the local shape/texture, the shape of the boundary, the overall spatial distribution of petals, and the colour. We combine the features using a multiple kernel framework with a SVM classifier. The weights for each class are learnt using the method of Varma and Ray, which has achieved state of the art performance on other large dataset, such as Caltech 101/256. Our dataset has a similar challenge in the number of classes, but with the added difficulty of large between class similarity and small within class similarity. Results show that learning the optimum kernel combination of multiple features vastly improves the performance, from 55.1% for the best single feature to 72.8% for the combination of all features.

@JournalArticle{M.2008, abstract = {We investigate to what extent combinations of features can improve classification performance on a large dataset of similar classes. To this end we introduce a 103 class flower dataset. We compute four different features for the flowers, each describing different aspects, namely the local shape/texture, the shape of the boundary, the overall spatial distribution of petals, and the colour. We combine the features using a multiple kernel framework with a SVM classifier. The weights for each class are learnt using the method of Varma and Ray, which has achieved state of the art performance on other large dataset, such as Caltech 101/256. Our dataset has a similar challenge in the number of classes, but with the added difficulty of large between class similarity and small within class similarity. Results show that learning the optimum kernel combination of multiple features vastly improves the performance, from 55.1% for the best single feature to 72.8% for the combination of all features.}, added-at = {2023-12-06T09:49:19.000+0100}, author = {Nilsback, M. and Zisserman, Andrew}, biburl = {https://www.bibsonomy.org/bibtex/289d8184c7fba61e2d11bc3079f29dbda/livie}, day = 16, description = {This paper explores the use of multiple features for improving classification performance in a large dataset of similar classes, specifically focusing on a 103 class flower dataset. It demonstrates significant improvement in classification accuracy by combining various features.}, doi = {10.1109/ICVGIP.2008.47}, interhash = {a39e54d3c32bf315479cc12bc2a30ae1}, intrahash = {89d8184c7fba61e2d11bc3079f29dbda}, journal = {2008 Sixth Indian Conference on Computer Vision, Graphics & Image Processing}, keywords = {flower_classification machine_learning computer_vision Oxford102_flower_dataset posted_with_chatgpt}, month = {12}, pages = {722-729}, timestamp = {2023-12-06T09:49:19.000+0100}, title = {Automated Flower Classification over a Large Number of Classes}, url = {https://www.semanticscholar.org/paper/02b28f3b71138a06e40dbd614abf8568420ae183}, year = 2008 }