摘要
Image classification with small amount of data has become a key issue in computer vision. In this paper, a novel stacking architecture based on support vector machine (SVM) and adaptive boosting (AdaBoost) is proposed. The 18-layer ResNet model is used to extract features, acting as the input of the stacking algorithm. Then, the SVM classifiers in the first layer produce the basic predictions, based on which, the AdaBoost classifier in the second layer decides the final class of image. Experimental results on CIFAR-10 dataset demonstrate that the SVM-AdaBoost stacking algorithm outperforms the existing competitive benchmark algorithms with single SVM or AdaBoost.
| 源语言 | 英语 |
|---|---|
| 主期刊名 | Proceedings of 2021 IEEE International Conference on Unmanned Systems, ICUS 2021 |
| 出版商 | Institute of Electrical and Electronics Engineers Inc. |
| 页 | 16-20 |
| 页数 | 5 |
| ISBN（电子版） | 9780738146577 |
| DOI | |
| 出版状态 | 已出版 - 2021 |
| 已对外发布 | 是 |
| 活动 | 2021 IEEE International Conference on Unmanned Systems, ICUS 2021 - Beijing, 中国 期限: 15 10月 2021 → 17 10月 2021 |
出版系列
| 姓名 | Proceedings of 2021 IEEE International Conference on Unmanned Systems, ICUS 2021 |
|---|
会议
| 会议 | 2021 IEEE International Conference on Unmanned Systems, ICUS 2021 |
|---|---|
| 国家/地区 | 中国 |
| 市 | Beijing |
| 时期 | 15/10/21 → 17/10/21 |
指纹
探究 'Improved SVM-AdaBoost Stacking Algorithm with ResNet18' 的科研主题。它们共同构成独一无二的指纹。引用此
Wei, M., Lv, Y., & Zhou, J. (2021). Improved SVM-AdaBoost Stacking Algorithm with ResNet18. 在 Proceedings of 2021 IEEE International Conference on Unmanned Systems, ICUS 2021 (页码 16-20). (Proceedings of 2021 IEEE International Conference on Unmanned Systems, ICUS 2021). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICUS52573.2021.9641106